自动化地面实况标记跨多个信号

% - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - %步骤1:定义所需的属性描述算法,%,包括名称、描述和UserDirections。属性(常数)%名称算法名称%特征向量算法指定的名称。Name = ' Multisignal车辆检测器;%描述算法描述%字符向量指定算法的简短描述。描述=[使用ACF车辆检测器检测车辆的……“形象和估计他们在点云。”);% % UserDirections算法使用方向单元阵列特征向量为%算法用户指定的方向。UserDirections ={(“选择一个矩形的ROI标签”……对象作为车辆的标签。'],……(点击设置和激光雷达相机校正的…… 'Parameters tab, load the cameraIntrinsics and rigid3d ' ... 'objects from the workspace.'], ... ['Specify additional parameters under Settings.'], ... ['Click Run to detect vehicles in each image and point cloud.'], ... ['Review automated labels manually. You can modify, delete ', ... 'and add new labels.'], ... ['If you are not satisfied with the results, click Undo ' ... 'Run. Click Settings to modify algorithm settings and click ', ... 'Run again.'] ... ['When you are satisfied with the results, click Accept and ', ... 'return to manual labeling.']}; end

% - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - %步骤2:定义属性被用来管理算法执行。属性% SelectedLabelName选中的标签名称%选择标签的名称。车辆检测算法将%被分配这个变量的名字。SelectedLabelName %探测器探测器% Pretrained车辆检测器,% acfObjectDetector类的一个对象。探测器% VehicleModelName车辆检测器模型名称% pretrained车辆探测器模型的名称。VehicleModelName =“众目睽睽”;% OverlapThreshold重叠阈值%阈值用于消除重叠边界框%在参考边界框,在0和1之间。%边界框重叠比率的分母,“RatioType”,设置为%“分钟”。OverlapThreshold = 0.45;分数阈值% % ScoreThreshold分类阈值用于拒绝检测检测%得分较低。 ScoreThreshold = 20; % ConfigureDetector Detection configuration flag % Boolean value that determines whether the detector is % configured using monoCamera sensor. ConfigureDetector = false; % SensorObj monoCamera sensor % Monocular camera sensor object, monoCamera, used to configure % the detector. A configured detector runs faster and can % potentially result in better detections. SensorObj = []; % SensorStr monoCamera sensor variable name % Character vector specifying the monoCamera object variable name % used to configure the detector. SensorStr = ''; % VehicleWidth Vehicle width % Vehicle width used to configure the detector, specified as % [minWidth, maxWidth], which describes the approximate width of the % object in world units. VehicleWidth = [1.5 2.5]; % VehicleLength Vehicle length % Vehicle length used to configure the detector, specified as % [minLength, maxLength] vector, which describes the approximate % length of the object in world units. VehicleLength = []; % IntrinsicsObj Camera intrinsics % cameraIntrinsics object, which represents a projective % transformation from camera to image coordinates. IntrinsicsObj = []; % IntrinsicsStr cameraIntrinsics variable name % cameraIntrinsics object variable name. IntrinsicsStr = ''; % ExtrinsicsObj Camera-to-lidar rigid transformation % rigid3d object representing the 3-D rigid geometric transformation % from the camera to the lidar. ExtrinsicsObj = []; % ExtrinsicsStr rigid3d variable name % Camera-to-lidar rigid3d object variable name. ExtrinsicsStr = ''; % ClusterThreshold Clustering threshold for two adjacent points % Threshold specifying the maximum distance between two adjacent points % for those points to belong to the same cluster. ClusterThreshold = 1; end

函数成功= supportsMulti万博1manbetxsignalAutomation支持MultiSignal (~) %。成功= true;结束

函数isValid = checkSignalType (signalType) %只有视频信号/图像序列和点云数据%是有效的。isValid =任何(signalType = = vision.labeler.loading.SignalType.Image) & &……任何(signalType = = vision.labeler.loading.SignalType.PointCloud);结束

函数isValid = checkLabelDefinition (~, labelDef) %只有矩形或长方体ROI标签定义为%车辆检测器是有效的。isValid = (labelDef。= = labelType类型。长方体||labelDef.Type == labelType.Rectangle); end

函数isReady = checkSetup (algObj ~) %有一个选定的ROI标签定义自动化吗?isReady = ~ isempty (algObj.SelectedLabelDefinitions);结束

函数settingsDialog (algObj) %调用对话框输入相机intrinsic和% camera-to-lidar刚性变换和选项选择% pretrained模型,重叠的阈值,检测分数%的阈值,和聚类阈值。可选地,输入%校准monoCamera传感器配置探测器。multiSignalVehicleDetectorSettings (algObj);结束

函数初始化(algObj ~) %存储选中的标签定义的名称。使用此%的名字标签检测到车辆。algObj。SelectedLabelName = algObj.SelectedLabelDefinitions.Name;%初始化车辆检测器pretrained模型。algObj。探测器= vehicleDetectorACF (algObj.VehicleModelName);结束

函数autoLabels =运行(algObj,我)% autoLabels单元阵列的长度一样的%信号的数量。autoLabels =细胞(大小(我,1),1);%得到图像的索引和PointCloud帧。如果isa(我{1},“pointCloud”) pcIdx = 1;imIdx = 2;其他imIdx = 1;pcIdx = 2;%检测边界框在图像帧结束。selectedBboxes = detectVehicle (algObj,我{imIdx, 1});%估算点云框架上边界框。 if ~isempty(selectedBboxes) % Store labels from the image. imageLabels = struct('Type', labelType.Rectangle, ... 'Name', algObj.SelectedLabelDefinitions.Name, ... 'Position', selectedBboxes); autoLabels{imIdx, 1} = imageLabels; % Remove the ground plane for the point cloud. groundPtsIndex = segmentGroundFromLidarData(I{pcIdx,1}, ... "ElevationAngleDelta", 15, "InitialElevationAngle", 10); nonGroundPts = select(I{pcIdx,1}, ~groundPtsIndex); % Predict 3-D bounding boxes. pcBboxes = bboxCameraToLidar(selectedBboxes, nonGroundPts, algObj.IntrinsicsObj, ... algObj.ExtrinsicsObj, "ClusterThreshold", algObj.ClusterThreshold); % Store labels from the point cloud. if(~isempty(pcBboxes)) pcLabels = struct('Type', labelType.Cuboid,... 'Name', algObj.SelectedLabelDefinitions.Name,... 'Position', pcBboxes); autoLabels{pcIdx, 1} = pcLabels; else autoLabels{pcIdx, 1} = {}; end else autoLabels{imIdx, 1} = {}; autoLabels{pcIdx, 1} = {}; end end

函数终止(~)

mkdir(“+视觉/ +贴标签机”);拷贝文件(“MultiSignalVehicleDetector.m”、“+视觉/ +贴标签机”);

imageURL = ' //www.tianjin-qmedu.com万博1manbetx/supportfiles/lidar/data/WPI_ImageData.tar.gz ';imageDataFolder = fullfile (tempdir WPI_ImageData, filesep);imageDataTarFile = imageDataFolder +“WPI_ImageData.tar.gz”;如果~存在(imageDataFolder“dir”) mkdir (imageDataFolder)如果~存在(imageDataTarFile,“文件”)disp (WPI形象驱动下载数据(225 MB)…”);websave (imageDataTarFile imageURL);解压(imageDataTarFile imageDataFolder);结束%焦油检查图像。gz文件下载,但不是未压缩的。如果存在~ (fullfile (imageDataFolder imageData),“dir”)解压(imageDataTarFile imageDataFolder)结束

%创建新的文件夹,复制图像。imDataFolder = imageDataFolder +“imageDataSequence”;如果~存在(imDataFolder“dir”) mkdir (imDataFolder);940年底我= 920:文件名= strcat (num2str(我,' % 06.0 f '), . jpg);源= fullfile (imageDataFolder imageData,文件名);目的地= fullfile (imageDataFolder, imageDataSequence,文件名);拷贝文件(来源、目的地)

lidarURL = ' //www.tianjin-qmedu.com万博1manbetx/supportfiles/lidar/data/WPI_LidarData.tar.gz ';lidarDataFolder = fullfile (tempdir WPI_LidarData, filesep);lidarDataTarFile = lidarDataFolder +“WPI_LidarData.tar.gz”;如果~存在(lidarDataFolder) mkdir (lidarDataFolder)如果~存在(lidarDataTarFile,“文件”)disp (WPI激光雷达驱动下载数据(760 MB)…”);websave (lidarDataTarFile lidarURL);解压(lidarDataTarFile lidarDataFolder);%检查激光雷达焦油。gz文件下载,但不是未压缩的。如果存在~ (fullfile (lidarDataFolder WPI_LidarData.mat),“文件”)解压(lidarDataTarFile lidarDataFolder);结束

%下载激光雷达数据加载到工作区中。负载(fullfile (lidarDataFolder WPI_LidarData.mat),“lidarData”);lidarData =重塑(lidarData、大小(lidarData, 2), 1);%创建新文件夹和激光雷达数据写入PCD文件。pcdDataFolder = lidarDataFolder +“lidarDataSequence”;如果~存在(pcdDataFolder“dir”) mkdir (fullfile (lidarDataFolder ' lidarDataSequence '));结束disp (PCD的驾驶WPI激光雷达数据保存文件…”);因为我= 920:940文件名= strcat (fullfile (lidarDataFolder, lidarDataSequence, filesep),…num2str(我,“% 06.0 f”)、“.pcd”);我pcwrite (lidarData{},文件名); end

data =负载(fullfile (toolboxdir(激光雷达),‘lidardata’,‘lcc’, ' bboxGT.mat '));cameraParams = data.cameraParams;camToLidar = data.camToLidar;

imageDir = fullfile (tempdir ' WPI_ImageData ', ' imageDataSequence ');pointCloudDir = fullfile (tempdir ' WPI_LidarData ', ' lidarDataSequence ');groundTruthLabeler