基于RCNN的汽车检测

%% 清除 clear;clc;close all; %% %% 加载数据 % vehicleDataset是一个dataset数据类型，第一列是图片的相对路径，第二列是图片中小汽车的位置 data = load('fasterRCNNVehicleTrainingData.mat'); % 提取训练集 vehicleDataset = data.vehicleTrainingData; % 提取图片路径 dataDir = fullfile(toolboxdir('vision'),'visiondata'); vehicleDataset.imageFilename = fullfile(dataDir, vehicleDataset.imageFilename); % 展示前9幅图片 k = 12; I=zeros(128,228,3,k); for i = 1 : k % 读取图片 tmp = imread(vehicleDataset.imageFilename{i}); % 添加标识框 tmp = insertShape(tmp, 'Rectangle', vehicleDataset.vehicle{i}); I(:,:,:,i) = mat2gray(tmp); end % 显示 Is = I; % hfig = figure; montage(Is); % set(hfig, 'Units', 'Normalized', 'Position', [0, 0, 1, 1]); % pause(1); %% 构建多输入多输出神经网络结构 deepNetworkDesigner % 添加层分支 layers = [ imageInputLayer([32 32 3],"Name","data") convolution2dLayer([3 3],96,"Name","conv1","BiasLearnRateFactor",2,"Stride",[2 2]) reluLayer("Name","relu1") crossChannelNormalizationLayer(5,"Name","norm1","K",1) maxPooling2dLayer([3 3],"Name","pool1","Stride",[2 2]) groupedConvolution2dLayer([5 5],128,2,"Name","conv2","BiasLearnRateFactor",2,"Padding",[2 2 2 2]) reluLayer("Name","relu2") crossChannelNormalizationLayer(5,"Name","norm2","K",1) maxPooling2dLayer([3 3],"Name","pool2","Stride",[2 2]) convolution2dLayer([3 3],384,"Name","conv3","BiasLearnRateFactor",2,"Padding",[1 1 1 1]) reluLayer("Name","relu3") groupedConvolution2dLayer([3 3],192,2,"Name","conv4","BiasLearnRateFactor",2,"Padding",[1 1 1 1]) reluLayer("Name","relu4") groupedConvolution2dLayer([3 3],128,2,"Name","conv5","BiasLearnRateFactor",2,"Padding",[1 1 1 1]) reluLayer("Name","relu5") maxPooling2dLayer([3 3],"Name","pool5","Stride",[2 2]) fullyConnectedLayer(4096,"Name","fc6","BiasLearnRateFactor",2) reluLayer("Name","relu6") dropoutLayer(0.5,"Name","drop6") fullyConnectedLayer(4096,"Name","fc7","BiasLearnRateFactor",2) reluLayer("Name","relu7") dropoutLayer(0.5,"Name","drop7") fullyConnectedLayer(2,"Name","fc8","BiasLearnRateFactor",2) softmaxLayer("Name","prob") classificationLayer("Name","output")]; % % 绘制层图 % plot(layerGraph(layers)); % 创建层图 lgraph = layers; % analyzeNetwork(layers) %% 训练CNN网络 % 将数据划分两部分 % 前60%的数据用于训练，后面40%用于测试 ind = round(size(vehicleDataset,1) * 0.6); trainData = vehicleDataset(1 : ind, :); testData = vehicleDataset(ind+1 : end, :); options =trainingOptions('sgdm', ... 'MiniBatchSize', 1, ... 'InitialLearnRate', 1e-3, ... 'LearnRateSchedule', 'piecewise', ... 'LearnRateDropFactor', 0.1, ... 'LearnRateDropPeriod', 100, ... 'MaxEpochs', 15, ... 'Verbose', true); % 设置模型的本地存储 doTrainingAndEval = 1; if doTrainingAndEval % 训练 R-CNN 神经网络，其实神经网络工具箱提供了3个函数 % （1）trainRCNNObjectDetector，训练快检测慢，允许指定proposalFcn % （2）trainFastRCNNObjectDetector，速度较快，允许指定proposalFcn % （3）trainFasterRCNNObjectDetector，优化运行性能，不需要指定proposalFcn detector = trainFasterRCNNObjectDetector(trainData, layers, options, ... 'NegativeOverlapRange', [0 0.3], ... 'PositiveOverlapRange', [0.6 1], ... 'BoxPyramidScale', 1.2); else % 加载已经训练好的神经网络 detector = data.detector; end I = imread('highway.png'); % 运行检测器，输出目标位置和得分 [bboxes, scores] = detect(detector, I); % 在图像上标记处识别的小汽车 I = insertObjectAnnotation(I, 'rectangle', bboxes, scores); figure; imshow(I) %% 评估训练效果 if doTrainingAndEval results = struct; for i = 1:size(testData,1) % 读取测试图片 I = imread(testData.imageFilename{i}); % 运行CNN检测器 [bboxes, scores, labels] = detect(detector, I); % 结果保存到结构体 results(i).Boxes = bboxes; results(i).Scores = scores; results(i).Labels = labels; end % 将结构体转换为table数据类型 results = struct2table(results); else % 加载之前评估好的数据 results = data.results; end % 从测试数据中提取期望的小车位置 expectedResults = testData(:, 2:end); %采用平均精确度评估检测效果 [ap, recall, precision] = evaluateDetectionPrecision(results, expectedResults); % 绘制召回率-精确率曲线 figure; plot(recall, precision); xlabel('Recall'); ylabel('Precision') grid on; title(sprintf('Average Precision = %.2f', ap));