#include "DetectPresenter.h" DetectPresenter::DetectPresenter(/* args */) { } DetectPresenter::~DetectPresenter() { } int DetectPresenter::DetectBag(std::vector& detectionDataCache, const SG_bagPositionParam& algoParam, const SSG_planeCalibPara& cameraCalibParam, const VrDebugParam& debugParam, LaserDataLoader& dataLoader, const double clibMatrix[16], DetectionResult& detectionResult) { if (detectionDataCache.empty()) { LOG_WARNING("No cached detection data available\n"); return ERR_CODE(DEV_DATA_INVALID); } if(debugParam.savePointCloud){ LOG_INFO("[Algo Thread] Debug mode is enabled\n"); // 获取当前时间戳,格式为YYYYMMDDHHMMSS std::string timeStamp = CVrDateUtils::GetNowTime(); std::string fileName = debugParam.debugOutputPath + "/pointCloud_" + timeStamp + ".txt"; dataLoader.SaveLaserScanData(fileName, detectionDataCache, detectionDataCache.size(), 0.0, 0, 0); } // 1. 使用成员变量算法参数(已在初始化时从XML读取) LOG_INFO("[Algo Thread] Using algorithm parameters from XML configuration\n"); LOG_INFO(" Bag: L=%.1f, W=%.1f, H=%.1f\n",algoParam.bagParam.bagL, algoParam.bagParam.bagW, algoParam.bagParam.bagH); LOG_INFO(" Filter: continuityTh=%.1f, outlierTh=%d\n", algoParam.filterParam.continuityTh, algoParam.filterParam.outlierTh); LOG_INFO(" Plane height: %.3f\n", cameraCalibParam.planeHeight); LOG_INFO(" Plane calibration matrix: [%.3f, %.3f, %.3f; %.3f, %.3f, %.3f; %.3f, %.3f, %.3f]\n", cameraCalibParam.planeCalib[0], cameraCalibParam.planeCalib[1], cameraCalibParam.planeCalib[2], cameraCalibParam.planeCalib[3], cameraCalibParam.planeCalib[4], cameraCalibParam.planeCalib[5], cameraCalibParam.planeCalib[6], cameraCalibParam.planeCalib[7], cameraCalibParam.planeCalib[8]); // 2. 数据预处理:调平和去除地面(使用当前相机的调平参数) for (size_t i = 0; i < detectionDataCache.size(); i++) { sg_lineDataR(&detectionDataCache[i], cameraCalibParam.planeCalib, cameraCalibParam.planeHeight); } LOG_INFO("[Algo Thread] clibMatrix: \n\t[%.3f, %.3f, %.3f, %.3f] \n\t[ %.3f, %.3f, %.3f, %.3f] \n\t[ %.3f, %.3f, %.3f, %.3f] \n\t[ %.3f, %.3f, %.3f, %.3f]\n", clibMatrix[0], clibMatrix[1], clibMatrix[2], clibMatrix[3], clibMatrix[4], clibMatrix[5], clibMatrix[6], clibMatrix[7], clibMatrix[8], clibMatrix[9], clibMatrix[10], clibMatrix[11], clibMatrix[12], clibMatrix[13], clibMatrix[14], clibMatrix[15]); // 3. 调用算法检测接口(使用当前相机的调平参数) std::vector objOps; sg_getBagPosition(static_cast(detectionDataCache.data()), detectionDataCache.size(), algoParam, cameraCalibParam, objOps); // 从点云数据生成投影图像 detectionResult.image = PointCloudImageUtils::GeneratePointCloudImage(static_cast(detectionDataCache.data()), detectionDataCache.size(), objOps); // 转换检测结果为UI显示格式(使用机械臂坐标系数据) for (size_t i = 0; i < objOps.size(); i++) { const SSG_peakRgnInfo& obj = objOps[i]; // 进行坐标转换:从算法坐标系转换到机械臂坐标系 SVzNL3DPoint targetObj; targetObj.x = obj.centerPos.x; targetObj.y = obj.centerPos.y; targetObj.z = obj.centerPos.z; SVzNL3DPoint robotObj; CVrConvert::EyeToRobot(targetObj, robotObj, clibMatrix); // 创建位置数据(使用转换后的机械臂坐标) GrabBagPosition pos; pos.x = robotObj.x; // 机械臂坐标X pos.y = robotObj.y; // 机械臂坐标Y pos.z = robotObj.z; // 机械臂坐标Z pos.roll = obj.centerPos.x_roll; // 保持原有姿态角 pos.pitch = obj.centerPos.y_pitch; // 保持原有姿态角 pos.yaw = obj.centerPos.z_yaw; // 保持原有偏航角 detectionResult.positions.push_back(pos); LOG_INFO("[Algo Thread] Object %zu Eye Coords: X=%.2f, Y=%.2f, Z=%.2f\n", i, obj.centerPos.x, obj.centerPos.y, obj.centerPos.z); LOG_INFO("[Algo Thread] Object %zu Robot Coords: X=%.2f, Y=%.2f, Z=%.2f, Roll=%.2f, Pitch=%.2f, Yaw=%.2f\n", i, pos.x, pos.y, pos.z, pos.roll, pos.pitch, pos.yaw); } return 0; } int DetectPresenter::DetectBag(std::vector& detectionDataCache, const SG_bagPositionParam& algoParam, const SSG_planeCalibPara& cameraCalibParam, const VrDebugParam& debugParam, LaserDataLoader& dataLoader, const double clibMatrix[16], const SSG_hsvCmpParam& hsvCmpParam, const RGB& rgbColorPattern, const double frontColorTemplate[RGN_HIST_SIZE], const double backColorTemplate[RGN_HIST_SIZE], DetectionResult& detectionResult) { if (detectionDataCache.empty()) { LOG_WARNING("No cached detection data available\n"); return ERR_CODE(DEV_DATA_INVALID); } if(debugParam.savePointCloud){ LOG_INFO("[Algo Thread] Debug mode is enabled\n"); // 获取当前时间戳,格式为YYYYMMDDHHMMSS std::string timeStamp = CVrDateUtils::GetNowTime(); std::string fileName = debugParam.debugOutputPath + "/pointCloud_" + timeStamp + ".txt"; dataLoader.SaveLaserScanData(fileName, detectionDataCache, detectionDataCache.size(), 0.0, 0, 0); } // 2. 数据预处理:调平和去除地面(使用当前相机的调平参数) for (size_t i = 0; i < detectionDataCache.size(); i++) { sg_lineDataR_RGBD(&detectionDataCache[i], cameraCalibParam.planeCalib, cameraCalibParam.planeHeight); } #if 0 LOG_INFO("[Algo Thread] clibMatrix: \n\t[%.3f, %.3f, %.3f, %.3f] \n\t[ %.3f, %.3f, %.3f, %.3f] \n\t[ %.3f, %.3f, %.3f, %.3f] \n\t[ %.3f, %.3f, %.3f, %.3f]\n", clibMatrix[0], clibMatrix[1], clibMatrix[2], clibMatrix[3], clibMatrix[4], clibMatrix[5], clibMatrix[6], clibMatrix[7], clibMatrix[8], clibMatrix[9], clibMatrix[10], clibMatrix[11], clibMatrix[12], clibMatrix[13], clibMatrix[14], clibMatrix[15]); // 1. 使用成员变量算法参数(已在初始化时从XML读取) LOG_INFO("[Algo Thread] Using algorithm parameters from XML configuration\n"); LOG_INFO(" Bag: L=%.1f, W=%.1f, H=%.1f\n",algoParam.bagParam.bagL, algoParam.bagParam.bagW, algoParam.bagParam.bagH); LOG_INFO(" Filter: continuityTh=%.1f, outlierTh=%.1f\n", algoParam.filterParam.continuityTh, algoParam.filterParam.outlierTh); LOG_INFO(" Corner: minEndingGap=%.1f, minEndingGap_z=%.1f, scale=%.1f, cornerTh=%.1f, jumpCornerTh_1=%.1f, jumpCornerTh_2=%.1f\n", algoParam.cornerParam.minEndingGap, algoParam.cornerParam.minEndingGap_z, algoParam.cornerParam.scale, algoParam.cornerParam.cornerTh, algoParam.cornerParam.jumpCornerTh_1, algoParam.cornerParam.jumpCornerTh_2); LOG_INFO(" Grow: minVTypeTreeLen=%.1f, minLTypeTreeLen=%.1f yDeviation_max=%.1f, zDeviation_max=%.1f, maxLineSkipNum=%d, maxSkipDistance=%.1f\n", algoParam.growParam.minVTypeTreeLen, algoParam.growParam.minLTypeTreeLen, algoParam.growParam.yDeviation_max, algoParam.growParam.zDeviation_max, algoParam.growParam.maxLineSkipNum, algoParam.growParam.maxSkipDistance); LOG_INFO(" Plane height: %.3f\n", cameraCalibParam.planeHeight); LOG_INFO(" Plane calibration matrix: [%.3f, %.3f, %.3f; %.3f, %.3f, %.3f; %.3f, %.3f, %.3f]\n", cameraCalibParam.planeCalib[0], cameraCalibParam.planeCalib[1], cameraCalibParam.planeCalib[2], cameraCalibParam.planeCalib[3], cameraCalibParam.planeCalib[4], cameraCalibParam.planeCalib[5], cameraCalibParam.planeCalib[6], cameraCalibParam.planeCalib[7], cameraCalibParam.planeCalib[8]); #endif // 3. 调用算法检测接口(使用当前相机的调平参数) std::vector objOps; int nRet = 0; sg_getBagPositionAndOrientation(static_cast(detectionDataCache.data()), detectionDataCache.size(), algoParam, cameraCalibParam, hsvCmpParam, rgbColorPattern, frontColorTemplate, backColorTemplate, objOps, &nRet); if(nRet != 0){ LOG_ERROR("sg_getBagPositionAndOrientation failed, error code: %d\n", nRet); return nRet; } // 从点云数据生成投影图像 detectionResult.image = PointCloudImageUtils::GeneratePointCloudImage( static_cast(detectionDataCache.data()), detectionDataCache.size(), objOps); // 转换检测结果为UI显示格式(使用机械臂坐标系数据) for (size_t i = 0; i < objOps.size(); i++) { const SSG_peakOrienRgnInfo& obj = objOps[i]; // 进行坐标转换:从算法坐标系转换到机械臂坐标系 SVzNL3DPoint targetObj; targetObj.x = obj.centerPos.x; targetObj.y = obj.centerPos.y; targetObj.z = obj.centerPos.z; SVzNL3DPoint robotObj; CVrConvert::EyeToRobot(targetObj, robotObj, clibMatrix); // 创建位置数据(使用转换后的机械臂坐标) GrabBagPosition pos; pos.x = robotObj.x; // 机械臂坐标X pos.y = robotObj.y; // 机械臂坐标Y pos.z = robotObj.z; // 机械臂坐标Z pos.roll = obj.centerPos.x_roll; // 保持原有姿态角 pos.pitch = obj.centerPos.y_pitch; // 保持原有姿态角 pos.yaw = obj.centerPos.z_yaw; // 保持原有偏航角 detectionResult.positions.push_back(pos); LOG_INFO("[Algo Thread] Object %zu Eye Coords: X=%.2f, Y=%.2f, Z=%.2f\n", i, obj.centerPos.x, obj.centerPos.y, obj.centerPos.z); LOG_INFO("[Algo Thread] Object %zu Robot Coords: X=%.2f, Y=%.2f, Z=%.2f, Roll=%.2f, Pitch=%.2f, Yaw=%.2f\n", i, pos.x, pos.y, pos.z, pos.roll, pos.pitch, pos.yaw); } return 0; }