diff --git a/.vscode/settings.json b/.vscode/settings.json index 0ae04505..f2df0341 100644 --- a/.vscode/settings.json +++ b/.vscode/settings.json @@ -63,6 +63,9 @@ "thread": "cpp", "cinttypes": "cpp", "typeindex": "cpp", - "typeinfo": "cpp" + "typeinfo": "cpp", + "cassert": "cpp", + "bitset": "cpp", + "regex": "cpp" } } \ No newline at end of file diff --git a/Cigarette/.vscode/settings.json b/Cigarette/.vscode/settings.json new file mode 100644 index 00000000..99035127 --- /dev/null +++ b/Cigarette/.vscode/settings.json @@ -0,0 +1,5 @@ +{ + "files.associations": { + "memory": "cpp" + } +} \ No newline at end of file diff --git a/Cigarette/Alg/YOLOManager.cpp b/Cigarette/Alg/YOLOManager.cpp new file mode 100644 index 00000000..9fc3c36e --- /dev/null +++ b/Cigarette/Alg/YOLOManager.cpp @@ -0,0 +1,36 @@ +//// YOLOManager.cpp +//#include "YOLOManager.h" +// +//YOLOManager::~YOLOManager() { +// // 这里可以添加清理资源的代码 +//} +#include "YOLOManager.h" +#include +#include + +bool YOLOManager::InitializeDetector(SysConf& sys_conf, uint8_t Num) { + + bool ret=true; + params[Num].rectConfidenceThreshold = sys_conf.ConfThreshold; + params[Num].modelPath = sys_conf.model_path.toStdString(); + params[Num].imgSize = { IMAGE_WIDTH, IMAGE_HEIGHT }; +#ifdef USE_CUDA + params[Num].cudaEnable = true; +#else + params[Num].cudaEnable = false; +#endif + //params[Num].modelType = YOLO_DETECT_V3; + params[Num].modelType = YOLO_DETECT_V8; + + if (params[Num].modelType == YOLO_DETECT_V8) + detector[Num] = std::unique_ptr(new YOLO_V8()); + else if (params[Num].modelType == YOLO_DETECT_V3) + detector[Num] = std::unique_ptr(new YOLO_V3()); + + if (detector[Num]) + { + ret&=detector[Num]->InitParam(params[Num]); + ret&=detector[Num]->CreateSession(params[Num].modelPath); + } + return ret; +} \ No newline at end of file diff --git a/Cigarette/Alg/YOLOManager.h b/Cigarette/Alg/YOLOManager.h new file mode 100644 index 00000000..4ad43913 --- /dev/null +++ b/Cigarette/Alg/YOLOManager.h @@ -0,0 +1,92 @@ +#pragma once +#include "common.h" +#include +#include +#include + +enum MODEL_TYPE +{ + YOLO_DETECT_V3 = 0, + //FLOAT32 MODEL + YOLO_DETECT_V8 = 1, + YOLO_POSE = 2, + YOLO_CLS = 3, + //FLOAT16 MODEL + YOLO_DETECT_V8_HALF = 4, + YOLO_POSE_V8_HALF = 5, + YOLO_CLS_HALF = 6 +}; + +typedef struct _DL_INIT_PARAM +{ + std::string modelPath; + MODEL_TYPE modelType = YOLO_DETECT_V8; + std::vector imgSize = { IMAGE_WIDTH, IMAGE_HEIGHT }; + float rectConfidenceThreshold = 0.6; + float iouThreshold = 0.5; + int keyPointsNum = 2;//Note:kpt number for pose + bool cudaEnable = false; + int logSeverityLevel = 3; + int intraOpNumThreads = 1; +} DL_INIT_PARAM; + + +typedef struct _DL_RESULT +{ + int classId; + float confidence; + cv::Rect box; + std::vector keyPoints; +} DL_RESULT; + + +class YOLO +{ +public: + virtual bool InitParam(DL_INIT_PARAM& iParams) = 0; + + virtual bool CreateSession(std::string model_path, std::string model_name) = 0; + virtual bool CreateSession(std::string model_path) = 0; + + virtual char* RunSession(cv::Mat& in, cv::Mat& out, std::vector >& results) = 0; + virtual char* RunSession(std::vector& vec_in, std::vector& vec_out, std::vector > >& vec_results) = 0; + + virtual char* WarmUpSession() = 0; + virtual char* WarmUpSession(int shoot) = 0; +}; + +class YOLOManager { +public: + // 获取单例实例的方法 + static YOLOManager& GetInstance() { + static YOLOManager instance; + return instance; + } + + // 获取 YOLO 检测器实例 + YOLO* GetDetector(uint8_t Num) const { + return detector[Num].get(); + } + + bool InitializeDetector(SysConf& sys_conf, uint8_t Num); + +private: + // 私有构造函数，防止外部实例化 + YOLOManager() = default; + + // 禁用拷贝构造函数和赋值操作符 + YOLOManager(const YOLOManager&) = delete; + YOLOManager& operator=(const YOLOManager&) = delete; + + ~YOLOManager() = default; + + // 初始化检测器 +#ifdef __DEBUG + DL_INIT_PARAM params[NumberOfSupportedCameras+1]; + std::unique_ptr detector[NumberOfSupportedCameras+1]; +#else + DL_INIT_PARAM params[NumberOfSupportedCameras]; + std::unique_ptr detector[NumberOfSupportedCameras]; +#endif +}; + diff --git a/Cigarette/alg_jd.cpp b/Cigarette/Alg/alg_jd_yolo3.cpp similarity index 79% rename from Cigarette/alg_jd.cpp rename to Cigarette/Alg/alg_jd_yolo3.cpp index e7025a19..4601fa80 100644 --- a/Cigarette/alg_jd.cpp +++ b/Cigarette/Alg/alg_jd_yolo3.cpp @@ -1,37 +1,38 @@ -#include "alg_jd.h" +#include "alg_jd_yolo3.h" #include //鎵闇鐨勫簱鏂囦欢 -extern SysConf g_sys_conf; -// Initialize the parameters -static float confThreshold = g_sys_conf.ConfThreshold * 0.01; // Confidence threshold -static float nmsThreshold = 0.4; // Non-maximum suppression threshold -static int inpWidth = 416; // Width of network's input image -static int inpHeight = 416; // Height of network's input image - -static std::vector classes; +bool YOLO_V3::InitParam(DL_INIT_PARAM& iParams) +{ + // Initialize the parameters + confThreshold = iParams.rectConfidenceThreshold * 0.01; // Confidence threshold + nmsThreshold = 0.4; // Non-maximum suppression threshold + inpWidth = 416; // Width of network's input image + inpHeight = 416; // Height of network's input image + return true; +} -bool AlgJd::init(QString model_path, QString model_name) +bool YOLO_V3::CreateSession(std::string model_path, std::string model_name) { // Load names of classes std::string classesFile; - cv::String modelConfiguration; - cv::String modelWeights; - QString image_path; + std::string modelConfiguration; + std::string modelWeights; + std::string image_path; if (model_path.length() > 0 && model_name.length() > 0) { // 鎷煎噾鐨勬ā鍨嬫枃浠惰矾寰 - modelWeights = model_path.toStdString() + "/" + model_name.toStdString(); - modelConfiguration = model_path.toStdString() + "/jd.cfg"; - classesFile = model_path.toStdString() + "/jd.names"; - image_path = model_path + "/" + "alg_jd.bmp"; + modelWeights = model_path + "/" + model_name; + modelConfiguration = model_path + "/jd.cfg"; + classesFile = model_path + "/jd.names"; + image_path = model_path + "/" + "alg_jd"+ std::string(SUFFIX); } else { modelWeights = "D:/model/jd.weights"; classesFile = "D:/model/jd.names"; // Give the configuration and weight files for the model modelConfiguration = "D:/model/jd.cfg"; - image_path = "D:/model/alg_jd.bmp"; + image_path = "D:/model/alg_jd"+ std::string(SUFFIX); } std::ifstream classNamesFile(classesFile.c_str()); @@ -48,57 +49,64 @@ bool AlgJd::init(QString model_path, QString model_name) // Load the network net = cv::dnn::readNetFromDarknet(modelConfiguration, modelWeights); +#ifdef USE_CUDA net.setPreferableBackend(cv::dnn::DNN_BACKEND_CUDA); net.setPreferableTarget(cv::dnn::DNN_TARGET_CUDA); +#endif - //cv::Mat image = cv::imread("alg_jd.bmp"); - cv::Mat image = cv::imread(image_path.toStdString()); + //cv::Mat image = cv::imread("alg_jd"+SUFFIX); + cv::Mat image = cv::imread(image_path); /*if (g_sys_conf.model_jpg_path.length() > 0) { image = cv::imread(g_sys_conf.model_jpg_path.toStdString()); } else { - image = cv::imread("D:/Release/alg_jd.bmp"); + image = cv::imread("D:/Release/alg_jd"+SUFFIX); }*/ //璇嗗埆涓寮犲浘锛屾祴璇曟ā鍨嬫槸鍚︽纭紝骞朵笖瀹屾垚GPU鏁版嵁鍔犺浇 if (!image.data) return false; //鍒ゆ柇娴嬭瘯鍥剧墖鏄惁姝ｅ父璇诲彇 std::vector > results; - detect(image, image, results); + RunSession(image, image, results); if (results.size() > 0) return true; //妫娴嬪埌鐩爣锛屽垯鍒濆鍖栨垚鍔 else return false; //鍚﹀垯鍒濆鍖栧け璐 } -bool AlgJd::test_detect() +bool YOLO_V3::CreateSession(std::string model_path) +{ + return CreateSession(model_path, ""); +} + +char* YOLO_V3::WarmUpSession() { cv::Mat m1; - m1 = cv::Mat(544, 728, CV_8UC3, cv::Scalar(0, 0, 0)); + m1 = cv::Mat(IMAGE_HEIGHT, IMAGE_WIDTH, CV_8UC3, cv::Scalar(0, 0, 0)); std::vector > results; double t = (double)cv::getTickCount(); - detect(m1, m1, results); + RunSession(m1, m1, results); t = ((double)cv::getTickCount() - t) / cv::getTickFrequency(); - DEBUG(" test_detect time process:%f\n", t); - return true; + DEBUG(" WarmUpSession time process:%f\n", t); + return (char*)"WarmUpSession pass"; } -bool AlgJd::test_detect_batcht(int shoot) +char* YOLO_V3::WarmUpSession(int shoot) { std::vector vec_in; std::vector vec_out; std::vector>> vec_results; cv::Mat m1, m2, m3; cv::Mat m4; - m1 = cv::Mat(544, 728, CV_8UC3, cv::Scalar(0, 0, 0)); - m2 = cv::Mat(544, 728, CV_8UC3, cv::Scalar(0, 0, 0)); - m3 = cv::Mat(544, 728, CV_8UC3, cv::Scalar(0, 0, 0)); - m4 = cv::Mat(544, 728, CV_8UC3, cv::Scalar(0, 0, 0)); + m1 = cv::Mat(IMAGE_HEIGHT, IMAGE_WIDTH, CV_8UC3, cv::Scalar(0, 0, 0)); + m2 = cv::Mat(IMAGE_HEIGHT, IMAGE_WIDTH, CV_8UC3, cv::Scalar(0, 0, 0)); + m3 = cv::Mat(IMAGE_HEIGHT, IMAGE_WIDTH, CV_8UC3, cv::Scalar(0, 0, 0)); + m4 = cv::Mat(IMAGE_HEIGHT, IMAGE_WIDTH, CV_8UC3, cv::Scalar(0, 0, 0)); double t = (double)cv::getTickCount(); switch (shoot) { case 1: { std::vector > results; - detect(m1, m1, results); + RunSession(m1, m1, results); break; } case 3:vec_in.push_back(m3); @@ -106,15 +114,15 @@ bool AlgJd::test_detect_batcht(int shoot) case 4:vec_in.push_back(m4); default: { vec_in.push_back(m1); - detect_batch(vec_in, vec_out, vec_results); + RunSession(vec_in, vec_out, vec_results); } } t = ((double)cv::getTickCount() - t) / cv::getTickFrequency(); - DEBUG(" test_detect_batcht time process:%f\n", t); - return true; + DEBUG(" WarmUpSession time process:%f\n", t); + return (char*)"WarmUpSession pass"; } -int AlgJd::detect(cv::Mat& _frame, cv::Mat& out, std::vector >& results) +char* YOLO_V3::RunSession(cv::Mat& _frame, cv::Mat& out, std::vector >& results) { cv::Mat frame = _frame.clone(); cv::Mat image_clone = frame.clone(); @@ -141,10 +149,11 @@ int AlgJd::detect(cv::Mat& _frame, cv::Mat& out, std::vector >& vec_results) +void YOLO_V3::analyse(cv::Mat vec_in, std::vector >& vec_results) { for (int i = 0; i < vec_results.size(); i++) { @@ -159,7 +168,7 @@ void AlgJd::analyse(cv::Mat vec_in, std::vector >& vec_ } // Get the names of the output layers -std::vector AlgJd::getOutputsNames(const cv::dnn::Net& net) +std::vector YOLO_V3::getOutputsNames(const cv::dnn::Net& net) { std::vector names; if (names.empty()) @@ -178,7 +187,7 @@ std::vector AlgJd::getOutputsNames(const cv::dnn::Net& net) return names; } -void AlgJd::post_process(cv::Mat& frame, std::vector& outs, std::vector >& results) +void YOLO_V3::post_process(cv::Mat& frame, std::vector& outs, std::vector >& results) { std::vector < std::vector> classIds(classes.size()); std::vector < std::vector> confidences(classes.size()); @@ -233,7 +242,7 @@ void AlgJd::post_process(cv::Mat& frame, std::vector& outs, std::vector } // Draw the predicted bounding box -void AlgJd::drawPred(int classId, float conf, int left, int top, int right, int bottom, cv::Mat& frame) +void YOLO_V3::drawPred(int classId, float conf, int left, int top, int right, int bottom, cv::Mat& frame) { cv::Scalar color; if (classId == 0) @@ -264,7 +273,7 @@ void AlgJd::drawPred(int classId, float conf, int left, int top, int right, int cv::putText(frame, label, cv::Point(left, top - 5), cv::FONT_HERSHEY_SIMPLEX, 0.8, color, 1); } -void AlgJd::detect_batch(std::vector& vec_in, std::vector& vec_out, std::vector > >& vec_results) +char* YOLO_V3::RunSession(std::vector& vec_in, std::vector& vec_out, std::vector > >& vec_results) { cv::Mat blobs; std::vector image; @@ -295,9 +304,10 @@ void AlgJd::detect_batch(std::vector& vec_in, std::vector& vec out = detectedFrame.clone(); vec_out.push_back(out); } + return (char*)"pass"; } -void AlgJd::post_process_batch(std::vector& vec_frame, std::vector& outs, std::vector > >& vec_results) +void YOLO_V3::post_process_batch(std::vector& vec_frame, std::vector& outs, std::vector > >& vec_results) { int batch = vec_frame.size(); double confidence;/// @@ -356,7 +366,7 @@ void AlgJd::post_process_batch(std::vector& vec_frame, std::vector g_sys_conf.ConfThreshold * 0.01)///璇嗗埆搴︿綆浜庨槇鍊糔G澶勭悊 + if (confidences[i][idx] > confThreshold)///璇嗗埆搴︿綆浜庨槇鍊糔G澶勭悊 { if (box.width > 15) {//璇嗗埆妗嗗搴﹀ぇ浜15鏄剧ず璇嗗埆锛屽皬浜庤涓烘棤鑳剁偣锛孨G澶勭悊 diff --git a/Cigarette/alg_jd.h b/Cigarette/Alg/alg_jd_yolo3.h similarity index 57% rename from Cigarette/alg_jd.h rename to Cigarette/Alg/alg_jd_yolo3.h index e3aa2ab2..c6c21bd9 100644 --- a/Cigarette/alg_jd.h +++ b/Cigarette/Alg/alg_jd_yolo3.h @@ -1,5 +1,4 @@ -#ifndef _CIGARETTE_JD -#define _CIGARETTE_JD +#pragma once #include #include @@ -8,18 +7,25 @@ #include #include #include -#include "common.h" -class AlgJd +#include "YOLOManager.h" + +class YOLO_V3:public YOLO { public: - bool init(QString model_path, QString model_name); - bool test_detect(); - bool test_detect_batcht(int shoot); - int detect(cv::Mat& in, cv::Mat& out, std::vector >& results); + bool InitParam(DL_INIT_PARAM& iParams)override; + + bool CreateSession(std::string model_path, std::string model_name)override; + bool CreateSession(std::string model_path)override; + + char* RunSession(cv::Mat& in, cv::Mat& out, std::vector >& results)override; + char* RunSession(std::vector& vec_in, std::vector& vec_out, std::vector > >& vec_results)override; + + char* WarmUpSession()override; + char* WarmUpSession(int shoot)override; + // Remove the bounding boxes with low confidence using non-maxima suppression void post_process(cv::Mat& frame, std::vector& outs, std::vector >& results); void CircleImagePro(cv::Mat src, cv::Mat dst, std::vector radius); - void detect_batch(std::vector& vec_in, std::vector& vec_out, std::vector > >& vec_results); // Remove the bounding boxes with low confidence using non-maxima suppression void post_process_batch(std::vector& vec_frame, std::vector& outs, std::vector > >& vec_results); void analyse(cv::Mat vec_in, std::vector >& vec_results); @@ -27,9 +33,13 @@ public: std::vector getOutputsNames(const cv::dnn::Net& net); // Draw the predicted bounding box void drawPred(int classId, float conf, int left, int top, int right, int bottom, cv::Mat& frame); + private: cv::dnn::Net net; std::vector classes; -}; -#endif //end of _CIGARETTE_JD + float confThreshold; // Confidence threshold + float nmsThreshold; // Non-maximum suppression threshold + int inpWidth; // Width of network's input image + int inpHeight; // Height of network's input image +}; diff --git a/Cigarette/Alg/alg_jd_yolo8.cpp b/Cigarette/Alg/alg_jd_yolo8.cpp new file mode 100644 index 00000000..912b3887 --- /dev/null +++ b/Cigarette/Alg/alg_jd_yolo8.cpp @@ -0,0 +1,442 @@ +#define _CRT_SECURE_NO_WARNINGS 1 +#include "alg_jd_yolo8.h" +#include +#include + +#define benchmark +#define min(a,b) (((a) < (b)) ? (a) : (b)) + +YOLO_V8::YOLO_V8() { + +} + + +YOLO_V8::~YOLO_V8() { + delete session; + for (char* name : inputNodeNames) { + delete[] name; + } + for (char* name : outputNodeNames) { + delete[] name; + } +} + +#ifdef USE_CUDA +namespace Ort +{ + template<> + struct TypeToTensorType { static constexpr ONNXTensorElementDataType type = ONNX_TENSOR_ELEMENT_DATA_TYPE_FLOAT16; }; +} +#endif + +template +char* BlobFromImage(cv::Mat& iImg, T* iBlob) { + int channels = iImg.channels(); + int imgHeight = iImg.rows; + int imgWidth = iImg.cols; + + for (int c = 0; c < channels; c++) { + for (int h = 0; h < imgHeight; h++) { + for (int w = 0; w < imgWidth; w++) { + iBlob[c * imgWidth * imgHeight + h * imgWidth + w] = T( + (iImg.at(h, w)[c]) / 255.0f); + } + } + } + return RET_OK; +} + + + +char* YOLO_V8::PreProcess(cv::Mat& iImg, std::vector iImgSize, cv::Mat& oImg) +{ + if (iImg.channels() == 3) + { + oImg = iImg.clone(); + cv::cvtColor(oImg, oImg, cv::COLOR_BGR2RGB); + } + else + { + cv::cvtColor(iImg, oImg, cv::COLOR_GRAY2RGB); + } + + switch (modelType) + { + case YOLO_DETECT_V8: + case YOLO_POSE: + case YOLO_DETECT_V8_HALF: + case YOLO_POSE_V8_HALF://LetterBox + { + if (iImg.cols >= iImg.rows) + { + resizeScales = iImg.cols / (float)iImgSize.at(0); + cv::resize(oImg, oImg, cv::Size(iImgSize.at(0), int(iImg.rows / resizeScales))); + } + else + { + resizeScales = iImg.rows / (float)iImgSize.at(0); + cv::resize(oImg, oImg, cv::Size(int(iImg.cols / resizeScales), iImgSize.at(1))); + } + cv::Mat tempImg = cv::Mat::zeros(iImgSize.at(0), iImgSize.at(1), CV_8UC3); + oImg.copyTo(tempImg(cv::Rect(0, 0, oImg.cols, oImg.rows))); + oImg = tempImg; + break; + } + case YOLO_CLS://CenterCrop + { + int h = iImg.rows; + int w = iImg.cols; + int m = min(h, w); + int top = (h - m) / 2; + int left = (w - m) / 2; + cv::resize(oImg(cv::Rect(left, top, m, m)), oImg, cv::Size(iImgSize.at(0), iImgSize.at(1))); + break; + } + } + return RET_OK; +} + +bool YOLO_V8::InitParam(DL_INIT_PARAM& iParams) +{ + // Initialize the parameters + rectConfidenceThreshold = iParams.rectConfidenceThreshold; + iouThreshold = iParams.iouThreshold; + imgSize = {640,640}; + modelType = iParams.modelType; + cudaEnable = iParams.cudaEnable; + intraOpNumThreads = iParams.intraOpNumThreads; + logSeverityLevel = iParams.logSeverityLevel; + return true; +} + +bool YOLO_V8::CreateSession(std::string model_path, std::string model_name) { + char* Ret = RET_OK; + // std::regex pattern("[\u4e00-\u9fa5]"); + // bool result = std::regex_search(model_path, pattern); + // if (result) { + // return false; + // } + model_path += "/jd.onnx"; + try { + env = Ort::Env(ORT_LOGGING_LEVEL_WARNING, "Yolo"); + Ort::SessionOptions sessionOption; + + if (cudaEnable) { + OrtCUDAProviderOptions cudaOption; + cudaOption.device_id = 0; + sessionOption.AppendExecutionProvider_CUDA(cudaOption); + } + + sessionOption.SetGraphOptimizationLevel(GraphOptimizationLevel::ORT_ENABLE_ALL); + sessionOption.SetIntraOpNumThreads(intraOpNumThreads); + sessionOption.SetLogSeverityLevel(logSeverityLevel); + +#ifdef _WIN32 + int ModelPathSize = MultiByteToWideChar(CP_UTF8, 0, model_path.c_str(), + static_cast(model_path.length()), nullptr, 0); + std::unique_ptr wide_cstr(new wchar_t[ModelPathSize + 1]); + MultiByteToWideChar(CP_UTF8, 0, model_path.c_str(), + static_cast(model_path.length()), wide_cstr.get(), ModelPathSize); + wide_cstr[ModelPathSize] = L'\0'; + const wchar_t* modelPath = wide_cstr.get(); +#else + const char* modelPath = model_path.c_str(); +#endif + + session = new Ort::Session(env, modelPath, sessionOption); + Ort::AllocatorWithDefaultOptions allocator; + + // 获取输入节点 + size_t inputNodesNum = session->GetInputCount(); + for (size_t i = 0; i < inputNodesNum; i++) { + Ort::AllocatedStringPtr input_node_name = session->GetInputNameAllocated(i, allocator); + // 动态分配足够大的空间 + size_t name_len = strlen(input_node_name.get()) + 1; + char* temp_buf = new char[name_len]; + strcpy(temp_buf, input_node_name.get()); + inputNodeNames.push_back(temp_buf); + } + + // 获取输出节点 + size_t OutputNodesNum = session->GetOutputCount(); + for (size_t i = 0; i < OutputNodesNum; i++) { + Ort::AllocatedStringPtr output_node_name = session->GetOutputNameAllocated(i, allocator); + // 动态分配足够大的空间 + size_t name_len = strlen(output_node_name.get()) + 1; + char* temp_buf = new char[name_len]; + strcpy(temp_buf, output_node_name.get()); + outputNodeNames.push_back(temp_buf); + } + + options = Ort::RunOptions{ nullptr }; + WarmUpSession(); + return true; + } + catch (const std::exception& e) { + std::string result = "[YOLO_V8]:" + std::string(e.what()); + char* merged = new char[result.length() + 1]; + std::strcpy(merged, result.c_str()); + std::cout << merged << std::endl; + delete[] merged; + return false; + } +} + +bool YOLO_V8::CreateSession(std::string model_path) +{ + return CreateSession(model_path, ""); +} + +//char* YOLO_V8::RunSession(cv::Mat& iImg, std::vector& oResult) { +char* YOLO_V8::RunSession(cv::Mat& in, cv::Mat& out, std::vector >& results){ + std::vector oResult; + out = in.clone(); + +#ifdef benchmark + clock_t starttime_1 = clock(); +#endif + + char* Ret = RET_OK; + cv::Mat processedImg; + PreProcess(in, imgSize, processedImg); + + if (modelType < 4) { + std::unique_ptr blob(new float[processedImg.total() * 3]); + BlobFromImage(processedImg, blob.get()); // 使用 .get() 获取原始指针 + std::vector inputNodeDims = { 1, 3, imgSize.at(0), imgSize.at(1) }; + TensorProcess(starttime_1, in, blob.get(), inputNodeDims, oResult); + } + else { +#ifdef USE_CUDA + std::unique_ptr blob(new half[processedImg.total() * 3]); + BlobFromImage(processedImg, blob.get()); // 使用 .get() 获取原始指针 + std::vector inputNodeDims = { 1, 3, imgSize.at(0), imgSize.at(1) }; + TensorProcess(starttime_1, in, blob.get(), inputNodeDims, oResult); +#endif + } + + // 绘制检测结果 + for (auto& re : oResult) { + cv::RNG rng(cv::getTickCount()); + //cv::Scalar color(rng.uniform(0, 256), rng.uniform(0, 256), rng.uniform(0, 256)); + cv::Scalar color(0, 255, 0); + cv::rectangle(out, re.box, color, 3); + + float confidence = floor(100 * re.confidence) / 100; + std::cout << std::fixed << std::setprecision(2); + std::string label = std::string("jd") + std::string(" ") + + std::to_string(confidence).substr(0, std::to_string(confidence).size() - 4); + + cv::rectangle( + out, + cv::Point(re.box.x, re.box.y - 25), + cv::Point(re.box.x + label.length() * 15, re.box.y), + color, + cv::FILLED + ); + + cv::putText( + out, + label, + cv::Point(re.box.x, re.box.y - 5), + cv::FONT_HERSHEY_SIMPLEX, + 0.75, + cv::Scalar(0, 0, 0), + 2 + ); + } + + if(!oResult.empty()) + { + for(uint16_t index=0;index& vec_in, std::vector& vec_out, std::vector > >& vec_results){ + std::vector oResult; + + return (char*)"RunSession Undefined"; +} + +template +char* YOLO_V8::TensorProcess(clock_t& starttime_1, cv::Mat& iImg, T* blob, + std::vector& inputNodeDims, + std::vector& oResult) { + // 创建输入张量 + Ort::Value inputTensor = Ort::Value::CreateTensor( + Ort::MemoryInfo::CreateCpu(OrtDeviceAllocator, OrtMemTypeCPU), + blob, + 3 * imgSize.at(0) * imgSize.at(1), + inputNodeDims.data(), inputNodeDims.size()); + +#ifdef benchmark + clock_t starttime_2 = clock(); +#endif + + // 运行推理 + auto outputTensor = session->Run(options, inputNodeNames.data(), &inputTensor, 1, + outputNodeNames.data(), outputNodeNames.size()); + +#ifdef benchmark + clock_t starttime_3 = clock(); +#endif + + // 检查输出是否有效 + if (outputTensor.empty()) { + return (char*)"[YOLO_V8]: No output from model"; + } + + Ort::TypeInfo typeInfo = outputTensor.front().GetTypeInfo(); + auto tensor_info = typeInfo.GetTensorTypeAndShapeInfo(); + std::vector outputNodeDims = tensor_info.GetShape(); + + // 根据模型类型处理输出 + switch (modelType) { + case YOLO_DETECT_V8: + case YOLO_DETECT_V8_HALF: { + // 检测模型的输出处理代码 + int signalResultNum = outputNodeDims[1];//84 + int strideNum = outputNodeDims[2];//8400 + std::vector class_ids; + std::vector confidences; + std::vector boxes; + cv::Mat rawData; + + if (modelType == YOLO_DETECT_V8) { + // FP32 + rawData = cv::Mat(signalResultNum, strideNum, CV_32F, outputTensor.front().GetTensorMutableData()); + } + else { + // FP16 + rawData = cv::Mat(signalResultNum, strideNum, CV_16F, outputTensor.front().GetTensorMutableData()); + rawData.convertTo(rawData, CV_32F); + } + + // 处理检测结果... + rawData = rawData.t(); + float* data = (float*)rawData.data; + + for (int i = 0; i < strideNum; ++i) { + float* classesScores = data + 4; + cv::Mat scores(1, this->classes.size(), CV_32FC1, classesScores); + cv::Point class_id; + double maxClassScore; + cv::minMaxLoc(scores, 0, &maxClassScore, 0, &class_id); + if (maxClassScore > rectConfidenceThreshold) { + confidences.push_back(maxClassScore); + class_ids.push_back(class_id.x); + float x = data[0]; + float y = data[1]; + float w = data[2]; + float h = data[3]; + + int left = int((x - 0.5 * w) * resizeScales); + int top = int((y - 0.5 * h) * resizeScales); + + int width = int(w * resizeScales); + int height = int(h * resizeScales); + + boxes.push_back(cv::Rect(left, top, width, height)); + } + data += signalResultNum; + } + + std::vector nmsResult; + cv::dnn::NMSBoxes(boxes, confidences, rectConfidenceThreshold, iouThreshold, nmsResult); + for (int i = 0; i < nmsResult.size(); ++i) { + int idx = nmsResult[i]; + DL_RESULT result; + result.classId = class_ids[idx]; + result.confidence = confidences[idx]; + result.box = boxes[idx]; + oResult.push_back(result); + } + break; + } + case YOLO_CLS: + case YOLO_CLS_HALF: { + // 分类模型的输出处理代码 + cv::Mat rawData; + if (modelType == YOLO_CLS) { + // FP32 + rawData = cv::Mat(1, this->classes.size(), CV_32F, outputTensor.front().GetTensorMutableData()); + } + else { + // FP16 + rawData = cv::Mat(1, this->classes.size(), CV_16F, outputTensor.front().GetTensorMutableData()); + rawData.convertTo(rawData, CV_32F); + } + + float* data = (float*)rawData.data; + DL_RESULT result; + for (int i = 0; i < this->classes.size(); i++) { + result.classId = i; + result.confidence = data[i]; + oResult.push_back(result); + } + break; + } + default: + std::cout << "[YOLO_V8]: " << "Not support model type." << std::endl; + } + +#ifdef benchmark + clock_t starttime_4 = clock(); + double pre_process_time = (double)(starttime_2 - starttime_1) / CLOCKS_PER_SEC * 1000; + double process_time = (double)(starttime_3 - starttime_2) / CLOCKS_PER_SEC * 1000; + double post_process_time = (double)(starttime_4 - starttime_3) / CLOCKS_PER_SEC * 1000; + // 输出性能信息... +#endif + + return RET_OK; +} + +template char* YOLO_V8::TensorProcess(clock_t&, cv::Mat&, float*, std::vector&, std::vector&); +#ifdef USE_CUDA +template char* YOLO_V8::TensorProcess(clock_t&, cv::Mat&, half*, std::vector&, std::vector&); +#endif + + +char* YOLO_V8::WarmUpSession() { + clock_t starttime_1 = clock(); + cv::Mat iImg = cv::Mat(cv::Size(imgSize.at(0), imgSize.at(1)), CV_8UC3); + cv::Mat processedImg; + PreProcess(iImg, imgSize, processedImg); + + if (modelType < 4) { + std::unique_ptr blob(new float[processedImg.total() * 3]); + BlobFromImage(processedImg, blob.get()); + std::vector inputNodeDims = { 1, 3, imgSize.at(0), imgSize.at(1) }; + Ort::Value input_tensor = Ort::Value::CreateTensor( + Ort::MemoryInfo::CreateCpu(OrtDeviceAllocator, OrtMemTypeCPU), + blob.get(), 3 * imgSize.at(0) * imgSize.at(1), + inputNodeDims.data(), inputNodeDims.size()); + auto output_tensors = session->Run(options, inputNodeNames.data(), &input_tensor, 1, + outputNodeNames.data(), outputNodeNames.size()); + } + else { +#ifdef USE_CUDA + std::unique_ptr blob(new half[processedImg.total() * 3]); + BlobFromImage(processedImg, blob.get()); + std::vector inputNodeDims = { 1, 3, imgSize.at(0), imgSize.at(1) }; + Ort::Value input_tensor = Ort::Value::CreateTensor( + Ort::MemoryInfo::CreateCpu(OrtDeviceAllocator, OrtMemTypeCPU), + blob.get(), 3 * imgSize.at(0) * imgSize.at(1), + inputNodeDims.data(), inputNodeDims.size()); + auto output_tensors = session->Run(options, inputNodeNames.data(), &input_tensor, 1, + outputNodeNames.data(), outputNodeNames.size()); +#endif + } + + return RET_OK; +} + +char* YOLO_V8::WarmUpSession(int shoot) +{ + return (char*)"WarmUpSession Undefined"; +} \ No newline at end of file diff --git a/Cigarette/Alg/alg_jd_yolo8.h b/Cigarette/Alg/alg_jd_yolo8.h new file mode 100644 index 00000000..967e4403 --- /dev/null +++ b/Cigarette/Alg/alg_jd_yolo8.h @@ -0,0 +1,67 @@ +#pragma once + +#define RET_OK nullptr + +#ifdef _WIN32 +#include +#include +#include +#endif + +#include +#include +#include +#include +#include +#include "YOLOManager.h" + +#ifdef USE_CUDA +#include +#endif + +class YOLO_V8 :public YOLO +{ +public: + YOLO_V8(); + + ~YOLO_V8(); + +public: + bool InitParam(DL_INIT_PARAM& iParams)override; + + bool CreateSession(std::string model_path, std::string model_name)override; + bool CreateSession(std::string model_path)override; + + char* RunSession(cv::Mat& in, cv::Mat& out, std::vector >& results)override; + char* RunSession(std::vector& vec_in, std::vector& vec_out, std::vector > >& vec_results)override; + + char* WarmUpSession()override; + char* WarmUpSession(int shoot)override; + + template + char* TensorProcess(clock_t& starttime_1, cv::Mat& iImg, T* blob, + std::vector& inputNodeDims, + std::vector& oResult); + + char* PreProcess(cv::Mat& iImg, std::vector iImgSize, cv::Mat& oImg); + + std::vector classes{}; + +private: + Ort::Env env; + Ort::Session* session; + bool cudaEnable; + Ort::RunOptions options; + //std::vector inputNodeNames; + //std::vector outputNodeNames; + int logSeverityLevel; + int intraOpNumThreads; + MODEL_TYPE modelType; + std::vector imgSize; + float rectConfidenceThreshold; + float iouThreshold; + float resizeScales;//letterbox scale + std::vector inputNodeNames; + std::vector outputNodeNames; +}; + diff --git a/Cigarette/Cigarette.vcxproj b/Cigarette/Cigarette.vcxproj index d3c6732b..8e751470 100644 --- a/Cigarette/Cigarette.vcxproj +++ b/Cigarette/Cigarette.vcxproj @@ -21,7 +21,7 @@ {B12702AD-ABFB-343A-A199-8E24837244A3} QtVS_v304 - 10.0.22621.0 + 10.0.26100.0 $(MSBuildProjectDirectory)\QtMsBuild @@ -137,7 +137,7 @@ UNICODE;_UNICODE;WIN32;WIN64;NOMINMAX;%(PreprocessorDefinitions) GeneratedFiles\$(ConfigurationName);GeneratedFiles;.\GeneratedFiles;.;.\GeneratedFiles\$(ConfigurationName);G:\code_library\c\opencv\4.3\build-opencv-cpu\include;$(ProjectDir)MvIMPACT;$(ProjectDir)Pylon6.2\include;$(ProjectDir)Common;$(ProjectDir)modbus;$(ProjectDir)MVS3.2.1\Include;$(ProjectDir)PLC;$(ProjectDir)Ui;$(ProjectDir)Thread;$(ProjectDir)3rd\include;%(AdditionalIncludeDirectories) - MaxSpeed + Disabled ProgramDatabase MultiThreadedDebugDLL true @@ -149,7 +149,7 @@ Console $(OutDir)\$(ProjectName).exe G:\code_library\c\opencv\4.3\build-opencv-cpu\x64\vc15\lib;$(ProjectDir)Pylon6.2\lib\Win64;$(ProjectDir)3rd\lib;%(AdditionalLibraryDirectories) - false + true opencv_world430d.lib;modbus.lib;mvDeviceManager.lib;MvCameraControl.lib;Qt5Mqttd.lib;%(AdditionalDependencies) @@ -211,23 +211,23 @@ UNICODE;_UNICODE;WIN32;WIN64;QT_NO_DEBUG;NDEBUG;NOMINMAX;WIN32_LEAN_AND_MEAN;%(PreprocessorDefinitions) - GeneratedFiles\$(ConfigurationName);GeneratedFiles;.\GeneratedFiles;.;.\GeneratedFiles\$(ConfigurationName);$(ProjectDir)MvIMPACT;$(ProjectDir)OpenCV455Simple\include;$(ProjectDir)Common;$(ProjectDir)Pylon6.2\include;$(ProjectDir)modbus;$(ProjectDir)MVS3.2.1\Include;$(ProjectDir)PLC;$(ProjectDir)Ui;$(ProjectDir)Thread;$(ProjectDir)3rd\include;%(AdditionalIncludeDirectories) + GeneratedFiles\$(ConfigurationName);GeneratedFiles;.\GeneratedFiles;.;.\GeneratedFiles\$(ConfigurationName);$(ProjectDir)MvIMPACT;$(ProjectDir)OpenCV455Simple\include;$(ProjectDir)Common;$(ProjectDir)Pylon6.2\include;$(ProjectDir)modbus;$(ProjectDir)MVS3.2.1\Include;$(ProjectDir)PLC;$(ProjectDir)Ui;$(ProjectDir)Thread;$(ProjectDir)3rd\include;$(ProjectDir)Alg;$(ProjectDir)ONNX\ONNX-cpu-1.13.1\include;%(AdditionalIncludeDirectories) ProgramDatabase MultiThreaded true true stdcpp14 - MaxSpeed + Disabled false - 4819;%(DisableSpecificWarnings) + 4819;4566;%(DisableSpecificWarnings) true Console $(OutDir)\$(ProjectName).exe - $(ProjectDir)OpenCV455Simple\win64\vc15\lib;$(ProjectDir)Pylon6.2\lib\Win64;$(ProjectDir)MvIMPACT\lib\win64;$(ProjectDir)MVS3.2.1\lib\win64;$(ProjectDir)modbus;$(ProjectDir)3rd\lib;%(AdditionalLibraryDirectories) - false - opencv_world455.lib;modbus.lib;mvDeviceManager.lib;MvCameraControl.lib;Qt5Mqtt.lib;%(AdditionalDependencies) + $(ProjectDir)OpenCV455Simple\win64\vc15\lib;$(ProjectDir)Pylon6.2\lib\Win64;$(ProjectDir)MvIMPACT\lib\win64;$(ProjectDir)MVS3.2.1\lib\win64;$(ProjectDir)modbus;$(ProjectDir)3rd\lib;$(ProjectDir)ONNX\ONNX-cpu-1.13.1\lib;%(AdditionalLibraryDirectories) + true + opencv_world455.lib;modbus.lib;mvDeviceManager.lib;MvCameraControl.lib;Qt5Mqtt.lib;onnxruntime.lib;%(AdditionalDependencies) @@ -253,7 +253,9 @@ - + + + @@ -306,6 +308,9 @@ + + + @@ -330,7 +335,6 @@ - diff --git a/Cigarette/Cigarette.vcxproj.filters b/Cigarette/Cigarette.vcxproj.filters index 9b906351..a3fe7122 100644 --- a/Cigarette/Cigarette.vcxproj.filters +++ b/Cigarette/Cigarette.vcxproj.filters @@ -45,6 +45,12 @@ {91e6e8fc-6df2-4732-839b-4f873694fc78} + + {65df07cd-a8ba-4d7b-bec8-75d2cc5b744a} + + + {488e44eb-1d85-49cc-83eb-08e589998bea} + @@ -59,9 +65,6 @@ Source Files - - Source Files - Source Files @@ -155,6 +158,15 @@ Source Files\Thread + + Source Files\Alg + + + Source Files\Alg + + + Source Files\Alg + @@ -234,9 +246,6 @@ Header Files - - Header Files - Header Files @@ -276,6 +285,15 @@ Header Files\Thread + + Header Files\Alg + + + Header Files\Alg + + + Header Files\Alg + diff --git a/Cigarette/Cigarette.vcxproj.user b/Cigarette/Cigarette.vcxproj.user index 0081915a..b75c5d87 100644 --- a/Cigarette/Cigarette.vcxproj.user +++ b/Cigarette/Cigarette.vcxproj.user @@ -10,8 +10,20 @@ 2024-10-04T06:46:34.3746998Z + + - 2024-10-04T06:46:35.5916473Z + 2025-09-08T06:21:32.5149335Z + + + + + + + + + + \ No newline at end of file diff --git a/Cigarette/Thread/CaptureThreadHIK.cpp b/Cigarette/Thread/CaptureThreadHIK.cpp index 1dc71de8..646bdb05 100644 --- a/Cigarette/Thread/CaptureThreadHIK.cpp +++ b/Cigarette/Thread/CaptureThreadHIK.cpp @@ -234,7 +234,7 @@ CaptureThreadHIK::CaptureThreadHIK(void* camhandle, bool boTerminated, int Num) //----------------------------------------------------------------------------- { p_unit_queue = new ASyncQueue(Unit_Queue_Size); - g_pImage_buf = (unsigned char*)malloc(3000 * 3000 * 3); + g_pImage_buf = (unsigned char*)malloc(2*(IMAGE_HEIGHT * IMAGE_WIDTH * 3));//涔2闃茬垎 } CaptureThreadHIK::~CaptureThreadHIK() { diff --git a/Cigarette/Thread/syncworkthread.cpp b/Cigarette/Thread/syncworkthread.cpp index 23b4034d..71e6c011 100644 --- a/Cigarette/Thread/syncworkthread.cpp +++ b/Cigarette/Thread/syncworkthread.cpp @@ -1,5 +1,5 @@ 锘#include "SyncWorkThread.h" -#include "alg_jd.h" +#include "YOLOManager.h" #include "common.h" #include "balluffcamera.h" #include "baslercamera.h" @@ -8,7 +8,6 @@ #include "exportData.h" #include -extern AlgJd alg_jd[NumberOfSupportedCameras]; //妫娴嬭兌鐐圭殑AI绠楁硶 extern ConfPath g_conf_path; extern SysConf g_sys_conf; //绯荤粺閰嶇疆鍙傛暟 extern DisplayLabelConf g_display_label_conf[NumberOfSupportedCameras]; @@ -70,7 +69,7 @@ void SyncWorkThread::run() local_g_image_sync_queue->take(element_vec); bool IsNGForAll = false; int merge_index; - cv::Mat merge_image = cv::Mat::zeros(512 * work_camera_nums, 640 * g_sys_conf.shoot[0], CV_8UC3); + cv::Mat merge_image = cv::Mat::zeros(IMAGE_HEIGHT * work_camera_nums, IMAGE_WIDTH * g_sys_conf.shoot[0], CV_8UC3); cv::Rect roi; int j = 0; // 瀹為檯宸ヤ綔鐨勭浉鏈烘爣璇嗭紝element_vec涓彲鑳芥湁鐩告満娌″湪宸ヤ綔 for (int i = 0; i < element_vec.size(); i++)//姣忎釜鐩告満鐨勫浘杞祦閬嶅巻 @@ -120,7 +119,7 @@ void SyncWorkThread::run() { cv::cvtColor(image, image, CV_BGR2RGB); //鐏板害鍥惧儚杞负褰╄壊鍥惧儚 } - cv::resize(image, image, cv::Size(640, 512 * unit_count)); + cv::resize(image, image, cv::Size(IMAGE_WIDTH, IMAGE_HEIGHT * unit_count)); if (local_SysConf.shoot[local_camera_number] == unit_count) { std::vector vec_in; @@ -145,12 +144,12 @@ void SyncWorkThread::run() std::vector> results; cv::Mat imagein, imageout; imagein = vec_in[0]; - alg_jd[local_camera_number].detect(imagein, imageout, results); + YOLOManager::GetInstance().GetDetector(local_camera_number)->RunSession(imagein, imageout, results); vec_out.push_back(imageout.clone()); vec_results.push_back(results); } else { - alg_jd[local_camera_number].detect_batch(vec_in, vec_out, vec_results); + YOLOManager::GetInstance().GetDetector(local_camera_number)->RunSession(vec_in, vec_out, vec_results); } QDateTime ts_jd = QDateTime::currentDateTime(); int time_process = ts_start.msecsTo(ts_jd); @@ -204,10 +203,10 @@ void SyncWorkThread::run() if (merge_index == work_camera_nums * unit_count) { file_name = g_conf_path.save_pics_path + "/" + now_ts.toString("yyyy-MM-dd") + "/" - + now_ts.toString("yyyy-MM-dd_HH-mm-ss_zzz_") + ".jpg"; + + now_ts.toString("yyyy-MM-dd_HH-mm-ss_zzz_") + SUFFIX; g_save_queue->put(std::make_pair(file_name.toLocal8Bit().constData(), merge_image)); #ifdef __TCPSend - QString sendName = now_ts.toString("yyyy-MM-dd_HH-mm-ss_zzz_") + ".jpg"; + QString sendName = now_ts.toString("yyyy-MM-dd_HH-mm-ss_zzz_") + SUFFIX; TCPSendInfo.pics_name = sendName.toLocal8Bit().constData(); TCP_Info_queue->put(TCPSendInfo); #endif @@ -279,11 +278,11 @@ void SyncWorkThread::run() + QString::number(local_camera_number + 1) + "/" + QString::number(index + 1) + "/" + now_ts.toString("yyyy-MM-dd_HH-mm-ss_zzz_") + QString::number(local_camera_number + 1) + "#" + "_" + QString::number(index + 1) + "_" + ng_reason_maps[ngReason] + - ".jpg"; + SUFFIX; QString remotePath = "/image/ng/" + now_ts.toString("yyyy-MM-dd_HH-mm-ss_zzz_") + QString::number(local_camera_number + 1) + "#" + "_" + QString::number(index + 1) + "_" + ng_reason_maps[ngReason] + - ".jpg"; + SUFFIX; g_save_queue->put(std::make_pair(file_name.toLocal8Bit().constData(), m)); m = vec_out[index].clone(); @@ -292,11 +291,11 @@ void SyncWorkThread::run() + QString::number(local_camera_number + 1) + "/" + QString::number(index + 1) + "/" + now_ts.toString("yyyy-MM-dd_HH-mm-ss_zzz_") + QString::number(local_camera_number + 1) + "#" + "_" + QString::number(index + 1) + "_" + ng_reason_maps[ngReason] + - ".jpg"; + SUFFIX; remotePath = "/image/ng_result/" + now_ts.toString("yyyy-MM-dd_HH-mm-ss_zzz_") + QString::number(local_camera_number + 1) + "#" + "_" + QString::number(index + 1) + "_" + ng_reason_maps[ngReason] + - ".jpg"; + SUFFIX; //g_save_queue->put(std::make_pair(file_name.toStdString(), m)); g_save_queue->put(std::make_pair(file_name.toLocal8Bit().constData(), m)); } diff --git a/Cigarette/Thread/threadReceive.cpp b/Cigarette/Thread/threadReceive.cpp index b77bca57..caf32b32 100644 --- a/Cigarette/Thread/threadReceive.cpp +++ b/Cigarette/Thread/threadReceive.cpp @@ -1,6 +1,8 @@ #include "threadReceive.h" #include "cigarette.h" #include "basecamera.h" +#include + void threadReceive::start_work() { //start(HighestPriority); diff --git a/Cigarette/Thread/workthread.cpp b/Cigarette/Thread/workthread.cpp index 72ff76b7..1c99999b 100644 --- a/Cigarette/Thread/workthread.cpp +++ b/Cigarette/Thread/workthread.cpp @@ -1,5 +1,5 @@ #include "workthread.h" -#include "alg_jd.h" +#include "YOLOManager.h" #include "common.h" #include "balluffcamera.h" #include "baslercamera.h" @@ -8,7 +8,6 @@ #include "exportData.h" #include -extern AlgJd alg_jd[NumberOfSupportedCameras]; //妫娴嬭兌鐐圭殑AI绠楁硶 extern ConfPath g_conf_path; extern SysConf g_sys_conf; //绯荤粺閰嶇疆鍙傛暟 extern DisplayLabelConf g_display_label_conf[NumberOfSupportedCameras]; @@ -128,12 +127,12 @@ void WorkThread::run() std::vector > results; cv::Mat imagein, imageout; imagein = vec_in[0]; - alg_jd[local_camera_number].detect(imagein, imageout, results); + YOLOManager::GetInstance().GetDetector(local_camera_number)->RunSession(imagein, imageout, results); vec_out.push_back(imageout.clone()); vec_results.push_back(results); } else { - alg_jd[local_camera_number].detect_batch(vec_in, vec_out, vec_results); + YOLOManager::GetInstance().GetDetector(local_camera_number)->RunSession(vec_in, vec_out, vec_results); } QDateTime ts_jd = QDateTime::currentDateTime(); int time_process = ts_start.msecsTo(ts_jd); @@ -185,12 +184,12 @@ void WorkThread::run() now_ts.toString("yyyy-MM-dd") + "/" + QString::number(local_camera_number + 1) + "/" + QString::number(index + 1) + "/" + now_ts.toString("yyyy-MM-dd_HH-mm-ss_zzz_") + QString::number(local_camera_number + 1) + - "#" + "_" + QString::number(index + 1) + ".jpg"; + "#" + "_" + QString::number(index + 1) + SUFFIX; std::string filename = file_name.toLocal8Bit().constData(); g_save_queue->put(std::make_pair(filename, m)); QString sendName = now_ts.toString("yyyy-MM-dd_HH-mm-ss_zzz_") + QString::number(local_camera_number + 1) + - "#" + "_" + QString::number(index + 1) + ".jpg"; + "#" + "_" + QString::number(index + 1) + SUFFIX; #ifdef __TCPSend TCPSendInfo.pics_name = sendName.toLocal8Bit().constData(); TCP_Info_queue->put(TCPSendInfo); @@ -262,11 +261,11 @@ void WorkThread::run() + QString::number(local_camera_number + 1) + "/" + QString::number(index + 1) + "/" + now_ts.toString("yyyy-MM-dd_HH-mm-ss_zzz_") + QString::number(local_camera_number + 1) + "#" + "_" + QString::number(index + 1) + "_" + ng_reason_maps[ngReason] + - ".jpg"; + SUFFIX; QString remotePath = "/image/ng/" + now_ts.toString("yyyy-MM-dd_HH-mm-ss_zzz_") + QString::number(local_camera_number + 1) + "#" + "_" + QString::number(index + 1) + "_" + ng_reason_maps[ngReason] + - ".jpg"; + SUFFIX; g_save_queue->put(std::make_pair(file_name.toLocal8Bit().constData(), m)); m = vec_out[index].clone(); @@ -275,11 +274,11 @@ void WorkThread::run() + QString::number(local_camera_number + 1) + "/" + QString::number(index + 1) + "/" + now_ts.toString("yyyy-MM-dd_HH-mm-ss_zzz_") + QString::number(local_camera_number + 1) + "#" + "_" + QString::number(index + 1) + "_" + ng_reason_maps[ngReason] + - ".jpg"; + SUFFIX; remotePath = "/image/ng_result/" + now_ts.toString("yyyy-MM-dd_HH-mm-ss_zzz_") + QString::number(local_camera_number + 1) + "#" + "_" + QString::number(index + 1) + "_" + ng_reason_maps[ngReason] + - ".jpg"; + SUFFIX; //g_save_queue->put(std::make_pair(file_name.toStdString(), m)); g_save_queue->put(std::make_pair(file_name.toLocal8Bit().constData(), m)); } diff --git a/Cigarette/alg_jd_ng.cpp b/Cigarette/alg_jd_ng.cpp deleted file mode 100644 index 67ad154e..00000000 --- a/Cigarette/alg_jd_ng.cpp +++ /dev/null @@ -1,517 +0,0 @@ -#include "alg_jd_ng.h" -#include //所需的库文件 - -extern SysConf g_sys_conf; - -// Initialize the parameters -static float confThreshold = 0.01; // Confidence threshold -static float nmsThreshold = 0.4; // Non-maximum suppression threshold -static int inpWidth = 416; // Width of network's input image -static int inpHeight = 416; // Height of network's input image - -static std::vector classes; - -bool AlgJd_ng::init_ng(QString model_path, QString model_name) -{ - // Load names of classes - std::string classesFile; - cv::String modelConfiguration; - cv::String modelWeights; - QString image_path; - - modelWeights = model_path.toStdString() + "/" + model_name.toStdString(); - modelConfiguration = model_path.toStdString() + "/jd_ng.cfg"; - classesFile = model_path.toStdString() + "/jd_ng.names"; - image_path = model_path + "/" + "alg_jd_ng.bmp"; - - std::ifstream classNamesFile(classesFile.c_str()); - if (classNamesFile.is_open()) - { - std::string className = ""; - while (std::getline(classNamesFile, className)) - classes.push_back(className); - } - else{ - return false; - } - - // Load the network - net_ng = cv::dnn::readNetFromDarknet(modelConfiguration, modelWeights); - net_ng.setPreferableBackend(cv::dnn::DNN_BACKEND_CUDA); - net_ng.setPreferableTarget(cv::dnn::DNN_TARGET_CUDA); - - //cv::Mat image = cv::imread("alg_jd.bmp"); - cv::Mat image = cv::imread(image_path.toStdString()); - - //识别一张图，测试模型是否正确，并且完成GPU数据加载 - if (!image.data) return false; //判断测试图片是否正常读取 - std::vector > results; - detect_ng(image, image, results); - if (results.size() > 0) - return true; //检测到目标，则初始化成功 - else - return false; //否则初始化失败 -} - -bool AlgJd_ng::test_detect_ng() -{ - cv::Mat m1; - m1 = cv::Mat(544, 728, CV_8UC3, cv::Scalar(0, 0, 0)); - std::vector > results; - double t = (double)cv::getTickCount(); - detect_ng(m1, m1, results); - t = ((double)cv::getTickCount() - t) / cv::getTickFrequency(); - DEBUG(" test_detect time process:%f\n", t); - return true; -} - -bool AlgJd_ng::test_detect_batcht_ng(int shoot) -{ - std::vector vec_in; - std::vector vec_out; - std::vector > > vec_results; - cv::Mat m1, m2, m3; - m1 = cv::Mat(544, 728, CV_8UC3, cv::Scalar(0, 0, 0)); - m2 = cv::Mat(544, 728, CV_8UC3, cv::Scalar(0, 0, 0)); - m3 = cv::Mat(544, 728, CV_8UC3, cv::Scalar(0, 0, 0)); - - double t = (double)cv::getTickCount(); - switch(shoot){ - case 1:{ - std::vector > results; - detect_ng(m1, m1, results); - break; - } - case 3:vec_in.push_back(m3); - case 2:vec_in.push_back(m2); - default:{ - vec_in.push_back(m1); - detect_batch_ng(vec_in, vec_out, vec_results); - } - } - t = ((double)cv::getTickCount() - t) / cv::getTickFrequency(); - DEBUG(" test_detect_batcht time process:%f\n", t); - return true; -} - -int AlgJd_ng::detect_ng(cv::Mat& _frame, cv::Mat &out, std::vector > &results) -{ - cv::Mat frame = _frame.clone(); - cv::Mat image_clone=frame.clone(); - // Process frames. - // Create a 4D blob from a frame. - cv::Mat blob; - cv::dnn::blobFromImage(frame, blob, 1/255.0, cv::Size(inpWidth, inpHeight), cv::Scalar(0,0,0), true, false); - - //Sets the input to the network - net_ng.setInput(blob); - - - // Runs the forward pass to get output of the output layers - std::vector outs; - net_ng.forward(outs, getOutputsNames_ng(net_ng)); - - - // Remove the bounding boxes with low confidence - post_process_ng(frame, outs, results); - - - // Write the frame with the detection boxes - cv::Mat detectedFrame; - frame.convertTo(detectedFrame, CV_8U); - //show detectedFrame - out=detectedFrame.clone(); - return results.size(); -} - -int AlgJd_ng::detect_ng(cv::Mat& _frame, cv::Mat& draw_frame, cv::Mat& out, std::vector>& results) -{ - cv::Mat frame = _frame.clone(); - - // Process frames. - // Create a 4D blob from a frame. - // 创建4D的blob用于网络输入 - cv::Mat blob; - cv::dnn::blobFromImage(frame, blob, 1 / 255.0, cv::Size(inpWidth, inpHeight), cv::Scalar(0, 0, 0), true, false); - - //Sets the input to the network - net_ng.setInput(blob); - - - // Runs the forward pass to get output of the output layers - std::vector outs; - net_ng.forward(outs, getOutputsNames_ng(net_ng)); - - - // Remove the bounding boxes with low confidence - // 删除低置信度的边界框 - post_process_ng(draw_frame, outs, results); - - // Write the frame with the detection boxes - cv::Mat detectedFrame; - draw_frame.convertTo(detectedFrame, CV_8U); - - //show detectedFrame - out = detectedFrame.clone(); - return results.size(); -} - -void AlgJd_ng::CircleImagePro_ng(cv::Mat src, cv::Mat dst, std::vector radius) { - QStringList TestData; - src.copyTo(dst); - float c = float(20.5) / float(68.9); - std::vector> contours; - // 图像预处理 - cv::Mat img; - cv::cvtColor(dst, img, cv::COLOR_BGR2GRAY); - GaussianBlur(img, img, cv::Size(3, 3), 1); - Canny(img, img, 50, 150, 3); - imshow("img", img); - - cv::Mat kernel = cv::getStructuringElement(cv::MORPH_RECT, cv::Size(3, 3)); // 获取卷积核 - dilate(img, img, kernel, cv::Point(-1, -1), 2); - findContours(img, contours, cv::RETR_EXTERNAL, cv::CHAIN_APPROX_SIMPLE); - img.release(); - - std::vector center(contours.size()); - std::vector rad(contours.size()); - if (contours.size() == 0) { - std::cout << "未找到检测物" << std::endl; - return; - } - for (size_t i = 0; i < contours.size(); ++i) { - minEnclosingCircle(contours[i], center[i], rad[i]); - radius.push_back(rad[i] * c); - } - - for (size_t j = 0; j < center.size(); ++j) { - std::string str = "radius: "; - std::string str1; - float r = 0.0; - circle(dst, center[j], rad[j], cv::Scalar(0, 0, 255), 1); - r = rad[j] * c; - str1 = format("%.4f", r); - str.insert(str.end(), str1.begin(), str1.end()); - TestData.append(QString::fromStdString(str)); - cv::imshow("dst", dst); - cv::putText(dst, str, center[j], 1, 2, cv::Scalar(0, 255, 0), 2, 8); - cv::waitKey(1); - } - TestData.clear(); -} - -void AlgJd_ng::analyse_ng(cv::Mat vec_in, std::vector>& vec_results) -{ - bool IsNG = false; - std::vector vec_jd_results; // 0 胶点 - std::vector vec_kz_results; // 1 卡纸距离 - - for (int i = 0; i < vec_results.size(); i++) - { - if (vec_results[i].first == 0) // jd - { - vec_jd_results.push_back(vec_results[i].second); - } - else if (vec_results[i].first == 1) // 卡纸距离 - { - vec_kz_results.push_back(vec_results[i].second); - } - } - std::sort(vec_jd_results.begin(), vec_jd_results.end(), sort_rect_by_y_center_ng);//升序排列，//jinhuan+ - - //jinhuan+ - if (vec_jd_results.size() && vec_kz_results.size()) - { - //从上往下第一个胶点的中心y值 - int top_jd_y = vec_jd_results[0].y + vec_jd_results[0].height / 2; - //从上往下最后一个胶点的中心y值 - int bottom_jd_y = vec_jd_results[vec_jd_results.size()-1].y + vec_jd_results[vec_jd_results.size()-1].height / 2; - //卡纸的上边沿y值 - int top_kz_y = vec_kz_results[0].tl().y; - //卡纸的下边沿y值 - int bottom_kz_y = vec_kz_results[0].br().y; - - //if(abs(top_jd_y- top_kz_y)>50);//50个像素 - //if (abs(bottom_kz_y - bottom_jd_y)>40);//40个像素 - } else { - IsNG = true;//卡纸和胶点数量不够 - } - //if (vec_kz_results.size() != 1) {//卡纸有褶皱 - // IsNG = true; - //} - - //jinhuan- - cv::waitKey(1); -} - -// Get the names of the output layers -std::vector AlgJd_ng::getOutputsNames_ng(const cv::dnn::Net& net) -{ - std::vector names; - if (names.empty()) - { - //Get the indices of the output layers, i.e. the layers with unconnected outputs - std::vector outLayers = net.getUnconnectedOutLayers(); - - //get the names of all the layers in the network - std::vector layersNames = net.getLayerNames(); - - // Get the names of the output layers in names - names.resize(outLayers.size()); - for (size_t i = 0; i < outLayers.size(); ++i) - names[i] = layersNames[outLayers[i] - 1]; - } - return names; -} - -void AlgJd_ng::post_process_ng(cv::Mat& frame, std::vector& outs, std::vector > &results) -{ - std::vector < std::vector> classIds(classes.size()); - std::vector < std::vector> confidences(classes.size()); - std::vector < std::vector> boxes(classes.size()); - for (size_t i = 0; i < outs.size(); ++i) - { - // Scan through all the bounding boxes output from the network and keep only the - // ones with high confidence scores. Assign the box's class label as the class - // with the highest score for the box. - // 得到每个输出的数据 - float* data = (float*)outs[i].data; - for (int j = 0; j < outs[i].rows; ++j, data += outs[i].cols) - { - // 得到每个输出的所有类别的分数 - cv::Mat scores = outs[i].row(j).colRange(5, outs[i].cols); - cv::Point classIdPoint; - double confidence; - // Get the value and location of the maximum score - cv::minMaxLoc(scores, 0, &confidence, 0, &classIdPoint); - - //if (confidence > g_sys_conf.ConfThresholds[classIdPoint.x]*0.0001) - if (confidence > confThreshold) - { - int centerX = (int)(data[0] * frame.cols); - int centerY = (int)(data[1] * frame.rows); - int width = (int)(data[2] * frame.cols); - int height = (int)(data[3] * frame.rows); - int left = centerX - width / 2; - int top = centerY - height / 2; - - classIds[classIdPoint.x].push_back(classIdPoint.x); - confidences[classIdPoint.x].push_back((float)confidence); - boxes[classIdPoint.x].push_back(cv::Rect(left, top, width, height)); - } - } - } - - - // Perform non maximum suppression to eliminate redundant overlapping boxes with - // lower confidences - - for (size_t i = 0; i < classes.size(); ++i) - { - std::vector indices; - cv::dnn::NMSBoxes(boxes[i], confidences[i], confThreshold, nmsThreshold, indices); - for (size_t j = 0; j < indices.size(); ++j) - { - int idx = indices[j]; - cv::Rect box = boxes[i][idx]; - //if (classIds[i][idx] == 2) continue; - //else if (classIds[i][idx] == 3) continue; - - drawPred_ng(classIds[i][idx], confidences[i][idx], box.x, box.y, - box.x + box.width, box.y + box.height, frame); - results.push_back(std::make_pair(classIds[i][idx], box)); - } - } -} - -// Draw the predicted bounding box -void AlgJd_ng::drawPred_ng(int classId, float conf, int left, int top, int right, int bottom, cv::Mat& frame) -{ - cv::Scalar color; - if (classId == 0) //jd - color = cv::Scalar(0, 0, 255); // 绿色 - else if (classId == 1) // kz - color = cv::Scalar(0, 255, 0); // - else if (classId == 2) // bm - color = cv::Scalar(0, 255, 0); - else if (classId == 3) // kzx - color = cv::Scalar(0, 255, 0); - else - color = cv::Scalar(255, 255, 255); // 白色 - - //Draw a rectangle displaying the bounding box - cv::rectangle(frame, cv::Point(left, top), cv::Point(right, bottom), color, 4); - - //Get the label for the class name and its confidence - std::string label = cv::format("%.2f%%", (conf*100));/// - if (!classes.empty()) - { - CV_Assert(classId < (int)classes.size()); - label = classes[classId] + ": " + label; - } - else - { - std::cout<<"classes is empty..."<& vec_in, std::vector &vec_out, std::vector>> &vec_results) -{ - cv::Mat blobs; - std::vector image; - - cv::dnn::blobFromImages(vec_in, blobs, 1 / 255.0, cv::Size(inpWidth, inpHeight), cv::Scalar(0, 0, 0), true, false); - - //Sets the input to the network - net_ng.setInput(blobs); - - // Runs the forward pass to get output of the output layers - std::vector outs; - net_ng.forward(outs, getOutputsNames_ng(net_ng)); - - for (int i = 0; i < vec_in.size(); i++) - { - image.push_back(vec_in[i].clone()); - } - - // Remove the bounding boxes with low confidence - post_process_batch_ng(image, outs, vec_results); - - - // Write the frame with the detection boxes - for (int i = 0; i < vec_in.size(); i++) - { - cv::Mat detectedFrame, out; - image[i].convertTo(detectedFrame, CV_8U); - out = detectedFrame.clone(); - vec_out.push_back(out); - } -} - -void AlgJd_ng::detect_batch_ng(std::vector& vec_in, std::vector& transits, std::vector& vec_out, std::vector > >& vec_results) -{ - cv::Mat blobs; - //std::vector image= transits; - - cv::dnn::blobFromImages(vec_in, blobs, 1 / 255.0, cv::Size(inpWidth, inpHeight), cv::Scalar(0, 0, 0), true, false); - - //Sets the input to the network - net_ng.setInput(blobs); - - // Runs the forward pass to get output of the output layers - std::vector outs; - net_ng.forward(outs, getOutputsNames_ng(net_ng)); - //for (int i = 0; i < vec_in.size(); i++) - //{ - // //image.push_back(vec_in[i].clone()); - // transits.push_back(vec_in[i].clone()); - //} - - // Remove the bounding boxes with low confidence - //post_process_batch_ng(image, outs, vec_results); - post_process_batch_ng(transits, outs, vec_results); - - // Write the frame with the detection boxes - for (int i = 0; i < vec_in.size(); i++) - { - cv::Mat detectedFrame, out; - //image[i].convertTo(detectedFrame, CV_8U); - transits[i].convertTo(detectedFrame, CV_8U); - out = detectedFrame.clone(); - vec_out.push_back(out); - } -} - -void AlgJd_ng::post_process_batch_ng(std::vector& vec_frame, std::vector& outs, std::vector > > &vec_results) -{ - int batch = vec_frame.size(); - double confidence;/// - for (int k = 0; k < batch; k++) - { - std::vector < std::vector> classIds(classes.size()); - std::vector < std::vector> confidences(classes.size()); - std::vector < std::vector> boxes(classes.size()); - - //std::cout << "outs.size()\t" << outs.size() << std::endl; - //std::cout << "Type\t" << outs[0].type() << std::endl; - - for (size_t i = 0; i < outs.size(); ++i) - { - //std::cout << "outs.dims\t" << outs[i].dims << std::endl; - //std::cout << "outs[" << i << "].rows\t" << outs[i].size[0] << std::endl; - //std::cout << "outs[" << i << "].cols\t" << outs[i].size[1] << std::endl; - //std::cout << "outs[" << i << "].cols\t" << outs[i].size[2] << std::endl; - // Scan through all the bounding boxes output from the network and keep only the - // ones with high confidence scores. Assign the box's class label as the class - // - cv::Mat m0(outs[i].size[1], outs[i].size[2], CV_32F, outs[i].data + outs[i].step[0] * k); - // with the highest score for the box. - float* data = (float*)m0.data; - for (int j = 0; j < m0.rows; ++j, data += m0.cols) - { - cv::Mat scores = m0.row(j).colRange(5, m0.cols); - cv::Point classIdPoint; - //double confidence; - // Get the value and location of the maximum score - cv::minMaxLoc(scores, 0, &confidence, 0, &classIdPoint); - - //if (confidence > g_sys_conf.ConfThresholds[classIdPoint.x]*0.0001) - if (confidence > confThreshold) - { - int centerX = (int)(data[0] * vec_frame[k].cols); - int centerY = (int)(data[1] * vec_frame[k].rows); - int width = (int)(data[2] * vec_frame[k].cols); - int height = (int)(data[3] * vec_frame[k].rows); - int left = centerX - width / 2; - int top = centerY - height / 2; - - classIds[classIdPoint.x].push_back(classIdPoint.x); - confidences[classIdPoint.x].push_back((float)confidence); - boxes[classIdPoint.x].push_back(cv::Rect(left, top, width, height)); - } - } - } - std::vector > results; - // Perform non maximum suppression to eliminate redundant overlapping boxes with - // lower confidences - for (size_t i = 0; i < classes.size(); ++i) - { - std::vector indices; - cv::dnn::NMSBoxes(boxes[i], confidences[i], confThreshold, nmsThreshold, indices); - for (size_t j = 0; j < indices.size(); ++j) - { - int idx = indices[j]; - cv::Rect box = boxes[i][idx]; - drawPred_ng(classIds[i][idx], confidences[i][idx], box.x, box.y, - box.x + box.width, box.y + box.height, vec_frame[k]); - results.push_back(std::make_pair(classIds[i][idx], box)); - } - } - vec_results.push_back(results); - } -} - -//jinhuan+ -bool sort_rect_by_x_center_ng(cv::Rect r1, cv::Rect r2) -{ - return (r1.x + r1.width / 2) < (r2.x + r2.width / 2); -} -bool sort_rect_by_y_center_ng(cv::Rect r1, cv::Rect r2) -{ - return (r1.y + r1.height / 2) < (r2.y + r2.height / 2); -} -bool sort_rect_by_height_ng(cv::Rect r1, cv::Rect r2) -{ - return (r1.height) < (r2.height); -} -bool sort_rect_by_width_ng(cv::Rect r1, cv::Rect r2) -{ - return (r1.width) < (r2.width); -} -//jinhuan- \ No newline at end of file diff --git a/Cigarette/alg_jd_ng.h b/Cigarette/alg_jd_ng.h deleted file mode 100644 index 58f8fd65..00000000 --- a/Cigarette/alg_jd_ng.h +++ /dev/null @@ -1,42 +0,0 @@ -#ifndef _CIGARETTE_JD_ng -#define _CIGARETTE_JD_ng - -#include -#include -#include -#include -#include -#include -#include -#include "common.h" -class AlgJd_ng -{ - public: - bool init_ng(QString model_path, QString model_name); - bool test_detect_ng(); - bool test_detect_batcht_ng(int shoot); - int detect_ng(cv::Mat& in, cv::Mat &draw_frame, cv::Mat &out, std::vector> &results); - int detect_ng(cv::Mat& in, cv::Mat &out, std::vector > &results); - // Remove the bounding boxes with low confidence using non-maxima suppression - void post_process_ng(cv::Mat& frame, std::vector& outs, std::vector> &results); - void CircleImagePro_ng(cv::Mat src, cv::Mat dst, std::vector radius); - void detect_batch_ng(std::vector& vec_in, std::vector &vec_out, std::vector > > &vec_results); - void detect_batch_ng(std::vector& vec_in, std::vector &transits, std::vector &vec_out, std::vector>> &vec_results); - // Remove the bounding boxes with low confidence using non-maxima suppression - void post_process_batch_ng(std::vector& vec_frame, std::vector& outs, std::vector>> &vec_results); - void analyse_ng(cv::Mat vec_in, std::vector > & vec_results); - // Get the names of the output layers - std::vector getOutputsNames_ng(const cv::dnn::Net& net); - // Draw the predicted bounding box - void drawPred_ng(int classId, float conf, int left, int top, int right, int bottom, cv::Mat& frame); - private: - cv::dnn::Net net_ng; - std::vector classes_ng; -}; -//jinhuan+ -bool sort_rect_by_x_center_ng(cv::Rect r1, cv::Rect r2); -bool sort_rect_by_y_center_ng(cv::Rect r1, cv::Rect r2); -bool sort_rect_by_height_ng(cv::Rect r1, cv::Rect r2); -bool sort_rect_by_width_ng(cv::Rect r1, cv::Rect r2); -//jinhuan- -#endif //end of _CIGARETTE_JD diff --git a/Cigarette/cigarette.cpp b/Cigarette/cigarette.cpp index 1919a784..a9b64b3f 100644 --- a/Cigarette/cigarette.cpp +++ b/Cigarette/cigarette.cpp @@ -8,6 +8,7 @@ #include #include "exportData.h" #include +#include int g_op_time; //鎿嶄綔鍛樻潈闄愭椂闀匡紙榛樿300绉掞級 int g_admin_time; //绠＄悊鍛樻搷浣滄潈闄愭椂闀匡紙榛樿300绉掞級 @@ -28,10 +29,6 @@ QDateTime g_ts_start; extern SingleCamInfoStruct SingleCamInfo[NumberOfSupportedCameras]; -AlgJd alg_jd[NumberOfSupportedCameras]; //妫娴嬭兌鐐圭殑AI绠楁硶 -#ifdef __DEBUG -AlgJd alg_test;//test AI绠楁硶 -#endif QThread* pThread[NumberOfSupportedCameras]; int work_camera_nums; @@ -105,7 +102,7 @@ VOID BeforeWork(int shoot[]) for (int i = 0; i < NumberOfSupportedCameras; i++) { if (SingleCamInfo[i].Detect && SingleCamInfo[i].IsOpen) { - alg_jd[i].test_detect_batcht(shoot[i]); + YOLOManager::GetInstance().GetDetector(i)->WarmUpSession(shoot[i]); } } } @@ -253,6 +250,14 @@ Cigarette::Cigarette(QWidget* parent) QString config_path = g_conf_path.config_path + "/conf.txt"; read_sys_config(g_sys_conf, config_path); //鍒濆鍖栫郴缁熼厤缃 + if (g_sys_conf.model_path.isEmpty()) { + g_sys_conf.model_path = "D:/model"; + } + + if (g_sys_conf.model_name.isEmpty()) { + g_sys_conf.model_name = "jd.weights"; + } + for (int i = 0; i < NumberOfSupportedCameras; i++) { #if defined(NumberOfSupportedCameras) && (NumberOfSupportedCameras > i) g_modbus_conf.kick[i] = 0; @@ -340,28 +345,18 @@ Cigarette::Cigarette(QWidget* parent) CreatWorkThread(0, 0, this); #endif +#ifdef __DEBUG + YOLOManager::GetInstance().InitializeDetector(g_sys_conf,NumberOfSupportedCameras); +#endif + //鍒涘缓鐩告満宸ヤ綔绾跨▼ for (int i = 0; i < NumberOfSupportedCameras; i++) { if (SingleCamInfo[i].Detect) { work_camera_nums++; cam_status_mat[i]->setStyleSheet(tr("background-color: rgb(0, 170, 0);")); - QString model_path, model_name; - if (g_sys_conf.model_path.isEmpty()) { - model_path = "D:/model"; - g_sys_conf.model_path = "D:/model"; - } - else - model_path = g_sys_conf.model_path; - - if (g_sys_conf.model_name.isEmpty()) { - model_name = "jd.weights"; - g_sys_conf.model_name = "jd.weights"; - } - else - model_name = g_sys_conf.model_name; - if (!alg_jd[i].init(model_path, model_name)) + if (!YOLOManager::GetInstance().InitializeDetector(g_sys_conf,i)) { QMessageBox::information(NULL, QStringLiteral("绯荤粺鑷澶辫触"), QStringLiteral("AI妯″瀷1鍒濆鍖栧け璐ワ紝璇锋鏌ョ▼搴忓畬鏁存"), QMessageBox::Ok); exit(-1); @@ -417,10 +412,6 @@ Cigarette::Cigarette(QWidget* parent) } -#ifdef __DEBUG - alg_test.init(g_sys_conf.model_path, g_sys_conf.model_name);//test AI绠楁硶 -#endif - //鑷姩鎵撳紑鎵鏈夌浉鏈 if (g_sys_conf.auto_open == 1) { @@ -856,14 +847,14 @@ void Cigarette::TestImg() } std::vector > results; cv::Mat output; - alg_test.detect(imagein, output, results); + YOLOManager::GetInstance().GetDetector(NumberOfSupportedCameras)->RunSession(imagein, output, results); std::string WindowName = "TestImg"; cv::namedWindow(WindowName, cv::WINDOW_NORMAL); cv::imshow(WindowName, output); cv::waitKeyEx(1); -#ifdef __ExportData - alg_test.analyse(imagein, results); -#endif +// #ifdef __ExportData +// alg_test.analyse(imagein, results); +// #endif } void Cigarette::TestImgs() @@ -893,15 +884,15 @@ void Cigarette::TestImgs() cv::Mat output; std::vector > results; - alg_test.detect(imagein, output, results); + YOLOManager::GetInstance().GetDetector(NumberOfSupportedCameras)->RunSession(imagein, output, results); std::string WindowName = "TestImg"; cv::namedWindow(WindowName, cv::WINDOW_NORMAL); cv::imshow(WindowName, output); int k = cv::waitKeyEx(1); if (k == 27)break;//ESC閿 -#ifdef __ExportData - alg_test.analyse(imagein, results); -#endif +// #ifdef __ExportData +// alg_test.analyse(imagein, results); +// #endif QCoreApplication::processEvents(); } } @@ -2983,6 +2974,8 @@ bool Cigarette::ControlCamOpenOrClose(int Num, bool OpenOrClose) ifc.binningHorizontal.write(2); ifc.binningVerticalMode.writeS("Sum"); ifc.binningVertical.write(2); + ifc.height.write(IMAGE_HEIGHT); + ifc.width.write(IMAGE_WIDTH); pFI[Num] = new FunctionInterface(pDev); if (pDev->interfaceLayout.isValid() && (pDev->interfaceLayout.read() == dilGenICam)) @@ -3120,6 +3113,9 @@ bool Cigarette::ControlCamOpenOrClose(int Num, bool OpenOrClose) baslerCamera->BinningHorizontal.SetValue(2); baslerCamera->BinningVertical.SetValue(2); + baslerCamera->Height.SetValue(IMAGE_HEIGHT); + baslerCamera->Width.SetValue(IMAGE_WIDTH); + pBaslerCaptureThread[Num] = new CaptureThreadBasler(baslerCamera, false, Num, g_sys_conf.shoot[Num]); #ifdef SYNC_CAMERA pBaslerCaptureThread[Num]->p_image_sync_queue = g_image_sync_queue; @@ -3251,6 +3247,12 @@ bool Cigarette::ControlCamOpenOrClose(int Num, bool OpenOrClose) nRet = MV_CC_SetFloatValue(camhandle, "Gain", g_sys_conf.gain[Num]); if (nRet) { std::cout << "can not set Gain" << std::endl; nnRet = nRet; } + nRet = MV_CC_SetHeight(camhandle, IMAGE_HEIGHT); + if (nRet) { std::cout << "can not MV_CC_SetHeight" << std::endl; nnRet = nRet; } + + nRet = MV_CC_SetWidth(camhandle, IMAGE_WIDTH); + if (nRet) { std::cout << "can not MV_CC_SetWidth" << std::endl; nnRet = nRet; } + pHIKCaptureThread[Num] = new CaptureThreadHIK(camhandle, false, Num); #ifdef SYNC_CAMERA pHIKCaptureThread[Num]->p_image_sync_queue = g_image_sync_queue; diff --git a/Cigarette/cigarette.h b/Cigarette/cigarette.h index fee9eeae..22698c98 100644 --- a/Cigarette/cigarette.h +++ b/Cigarette/cigarette.h @@ -6,7 +6,7 @@ #include "dialogin.hpp" #include "dialogsetup.hpp" #include "plcsetup.hpp" -#include "alg_jd.h" +#include "YOLOManager.h" #include "balluffcamera.h" #include "baslercamera.h" #include "hikcamera.h" diff --git a/Cigarette/common.h b/Cigarette/common.h index e8fc5e9e..a911e243 100644 --- a/Cigarette/common.h +++ b/Cigarette/common.h @@ -5,6 +5,7 @@ #include "basecamera.h" #include "QtCore\qdatetime.h" +#define USE_CUDA //#define __DEBUG //debug淇℃伅杈撳嚭鍔熻兘 #define LOG_RECORD //log鏃ュ織鍔熻兘 #ifdef LOG_RECORD @@ -64,11 +65,15 @@ #define STATISTIC_FILE "camera%1_statistic.txt" #define ALARM_RECORD_FILE "alarm.txt" #define OUTPUT_HIGH_WIDTH 20000 //杈撳嚭淇″彿鐨勮剦鍐插搴︼紝寰 +#define SUFFIX ".jpg" #define OP_TIME 300 //OP鏉冮檺鏃堕暱锛堥粯璁300绉掞級 #define ADMIN_TIME 600 //ADMIN鏉冮檺鏃堕暱锛堥粯璁300绉掞級 #define STOP_SECONDS 3 //妫鏌ュ灏戞涓嶅彉瑙﹀彂鑷姩鎹㈢彮 +#define IMAGE_HEIGHT 512 +#define IMAGE_WIDTH 640 + int string_split(std::string str, std::string pattern, std::vector& out); std::string format(const char* pszFmt, ...); void getFiles(std::string path, std::vector& files); diff --git a/CigaretteSingle/CigaretteSingle.vcxproj.user b/CigaretteSingle/CigaretteSingle.vcxproj.user index 8c93fb0c..ed5ab64c 100644 --- a/CigaretteSingle/CigaretteSingle.vcxproj.user +++ b/CigaretteSingle/CigaretteSingle.vcxproj.user @@ -3,8 +3,12 @@ 2024-04-07T16:40:53.3929255Z + + 2023-05-12T00:20:07.0175398Z + + \ No newline at end of file