加载具有地面真实数据的MAT文件。加载=加载“caltech_cordova1_laneAndVehicleGroundTruth.mat”);

创建一个VideoReader对象来读取视频的帧。videoName = loaded.videoName;fileReader = VideoReader(videoName);地面真实数据按时间表组织。gtdata = loaded.groundTruthData;%显示地面真值数据的前几行。头(gtdata)

阅读视频的第一帧。frame = readFrame(fileReader);提取第一帧中的所有车道点。lanePoints = gtdata.LaneBoundaries{1};在第一帧中提取车辆边界框。vehicleBBox = gtdata.Vehicles{1};叠加右侧车道点和车辆边界框。frame = insertMarker(frame, lanePoints{2}，“X”);frame = insertobjectnotation(帧，“矩形”vehicleBBox,“汽车”);在第一帧显示地面真实数据。图imshow(框架)

设置monoSensorHelper来处理视频。monocamerassensor =加载的传感器;monoSensorHelper = helperMonoSensor(monocamerassensor);%创建新的时间表，使用相同的时间向量进行测量。测量=时间表(gtdata.Time);%设置保存车道边界和车辆数据的时间表栏。numFrames = floor(fileReader.FrameRate*fileReader.Duration);测量。LaneBoundaries = cell(numFrames, 2);测量。VehicleDetections = cell(numFrames, 1);gtdata。LanesInVehicleCoord = cell(numFrames, 2);将视频倒带到t = 0，并创建一个帧索引来保持当前状态%框架。fileReader。CurrentTime = 0;frameIndex = 0;循环播放视频文件，直到没有新的帧。而hasFrame(fileReader) frameIndex = frameIndex+1;frame = readFrame(fileReader);使用processFrame方法来计算检测。此方法可替换为自定义车道检测方法。detection = processFrame(monoSensorHelper, frame);存储估计的车道边界和车辆检测。测量。LaneBoundaries{frameIndex} = [detection . leftegoboundary .…detections.rightEgoBoundary];测量。VehicleDetections{frameIndex} = detection . vehicleboxes;为便于比较，将地面真理巷点转换为%车辆坐标系统。gtPointsThisFrame = gtdata.LaneBoundaries{frameIndex};vehiclePoints = cell(1, numel(gtPointsThisFrame));为ii = 1: nummel (gtPointsThisFrame) vehiclePoints{ii} = imageToVehicle(monocamerassensor, gtPointsThisFrame{ii});结束%存储以车辆坐标表示的地面真实点。gtdata。LanesInVehicleCoord{frameIndex} = vehiclePoints;结束

%将视频倒带到t = 0。fileReader。CurrentTime = 0;阅读视频的第一帧。frame = readFrame(fileReader);birdsEyeImage = transformImage(monoSensorHelper)。BirdsEyeConfig,框架);在鸟瞰图中提取第一帧的右巷点。firstFrameVehiclePoints = gtdata.LanesInVehicleCoord{1};pointsInBEV = vehicleToImage(monoSensorHelper)。BirdsEyeConfig, firstFrameVehiclePoints {2});在框架上叠加点。birdsEyeImage = insertMarker(birdsEyeImage, pointsInBEV，“X”，“大小”6);在鸟瞰中显示变换后的点。图imshow (birdsEyeImage)

阈值= 0.25;米百分比[numMatches, numMisses, numfalse positive, assignments] =…evaluateLaneBoundaries(测量。LaneBoundaries,…gtdata。LanesInVehicleCoord,…阈值);disp ([“火柴数:”num2str (numMatches)]);disp ([“失手次数:”num2str (numMisses)]);disp ([“误报次数:”num2str (numFalsePositives)]);

averageDistance = helpercomputelanstatistics(测量)。LaneBoundaries,…gtdata。LanesInVehicleCoord,…作业,@mean);%估算值与地面真实值之间的平均距离。图阻止(gtdata。时间,averageDistance) title(“估算值与实际情况之间的平均距离”网格)在ylabel (“米距离”)传说(的左边界，“右边界”）

使用单相机计算估计的点位置。[estVehiclePoints, estImagePoints] = helperGetCorrespondingPoints(monocamerassensor)，…测量。LaneBoundaries,…gtdata。LanesInVehicleCoord,…作业);在测量时间表中添加估计车道。测量。EstimatedLanes = estImagePoints;测量。LanesInVehicleCoord = estVehiclePoints;创建一个包含可视化所需的所有变量的新时间表。名称= {“LanePoints”；“DetectedLanePoints”};type = labelType({“行”；“行”});labelDefs = table(名称，类型，“VariableNames”， {“名字”，“类型”});visualizeInFrame =时间表时间,…gtdata。LaneBoundaries,…测量。EstimatedLanes,…“VariableNames”、名称);%创建groundTruth对象。dataSource = groundTruthDataSource(videoName);dataToVisualize = groundTruth(dataSource, labeldef, visualizeInFrame);%将前一节的所有结果保存在distanceData中。在一个%临时文件夹。dataToLoad = [tempdir . dataToLoad“distanceData.mat”];保存(dataToLoad,“monoSensorHelper”，“videoName”，“测量”，“gtdata”，“averageDistance”);

函数stat = helperComputeLaneStatistics(estModels, gtPoints, assignments, fcnHandle);默认情况下，左估计和右估计NaN表示缺乏%的数据。stat = NaN*ones(numFrames, 2);为frameInd = 1:numFrames%默认情况下进行左右估计NaN。stat(frameInd，:) = NaN*ones(2,1);为idx = 1:长度(estModels{frameInd})忽略假阳性赋值。如果赋值{frameInd}(idx) == 0继续；结束estModelInFrame中的第k个边界匹配到第k个%的元素被gtPointsInFrame中的赋值索引。thisModel = estModels{frameInd}(idx);thisGT = gtPoints{frameInd}{assignments{frameInd}(idx)};thisGTModel = driving.internal.piecewiseLinearBoundary(thisGT);如果mean(thisGTModel.Points(:，2)) > 0%左车道xPoints = thisGTModel.Points(，1);yDist = 0 (size(xPoints));为索引= 1:nummel (xPoints) gtYPoints = thisGTModel.computeBoundaryModel(xPoints(index));testYPoints = thisModel.computeBoundaryModel(xPoints(index));yDist(index) = abs(testYPoints-gtYPoints);结束stat(frameInd, 1) = fcnHandle(yDist);其他的%右车道xPoints = thisGTModel.Points(，1);yDist = 0 (size(xPoints));为索引= 1:nummel (xPoints) gtYPoints = thisGTModel.computeBoundaryModel(xPoints(index));testYPoints = thisModel.computeBoundaryModel(xPoints(index));yDist(index) = abs(testYPoints-gtYPoints);结束stat(frameInd, 2) = fcnHandle(yDist);结束结束结束结束

函数[vehiclePoints, imagePoints] = helperGetCorrespondingPoints(monocamerassensor, estModels, gtPoints, assignments) numFrames = length(estModels);imagePoints = cell(numFrames, 1);vehiclePoints = cell(numFrames, 1);为frameInd = 1:numFrames如果isempty(assignments{frameInd}) imagePointsInFrame = [];vehiclePointsInFrame = [];其他的estModelInFrame = estModels{frameInd};gtPointsInFrame = gtPoints{frame};imagePointsInFrame = cell(length(estModelInFrame)， 1);vehiclePointsInFrame = cell(length(estModelInFrame)， 1);为idx = 1:length(estModelInFrame)忽略假阳性赋值。如果赋值{frameInd}(idx) == 0 imagePointsInFrame{idx} = [NaN NaN];继续；结束estModelInFrame中的第k个边界匹配到第k个%的元素被gtPointsInFrame中的赋值索引。thisModel = estModelInFrame(idx);thisGT = gtPointsInFrame{assignments{frameInd}(idx)};xPoints = thisGT(:， 1);yPoints = thisModel.computeBoundaryModel(xPoints);vehiclePointsInFrame{idx} = [xPoints, yPoints];imagePointsInFrame{idx} = vehicleToImage(monocamerassensor， [xPoints yPoints]);结束结束车辆点{frameInd} =车辆点sinframe;imagePoints{frameInd} = imagePointsInFrame;%使imagePoints[]而不是{}符合groundTruth对象。如果isempty(imagePoints{frameInd}) imagePoints{frameInd} = []结束如果isempty(车辆点{frameInd})车辆点{frameInd} = [];结束结束结束