faster rcnn matlab,Faster rcnn 模型的数据标定的知识求助，matlab2017b自带的函数

%构建CNN网络

%输入层，最小检测对象设置为540*960

inputLayer = imageInputLayer([10 10 3]);

%中间层

%定义卷基层参数

filterSize = [3 3];

numFilters =32;

middleLayers = [

%第一轮，只包含CNN和ReLU

convolution2dLayer(filterSize,numFilters,'Padding',1)

reluLayer()

%第二轮，包含CNN、ReLU和Pooling

convolution2dLayer(filterSize,numFilters,'Padding',1)

reluLayer()

maxPooling2dLayer(3,'Stride',2)

];

%输出层

finalLayers = [

%新增一个包含64个输出的全连接层

fullyConnectedLayer(64)

%新增一个非线性ReLU层

reluLayer()

%新增一个2个输出的全连接层，用于判断图像是否半酣检测对象

fullyConnectedLayer(2)

%添加softmax和classification层

softmaxLayer()

classificationLayer()

];

%组合所有层

layers=[

inputLayer

middleLayers

finalLayers

];

%%训练CNN网络

%将数据划分两部分

%前60%数据用于训练，后面40%用于测试

ind = round(size(data1,1)*0.6);

trainData = data1(1:ind,:);

testData = data1(ind+1:end,:);

%训练过程包括4步，每步可以使用单独的参数，与可以使用同一个参数

options=[

%第1步，Training a Region Proposal Network (RPN)

trainingOptions('sgdm','Maxepochs',10,'InitialLearnRate',1e-5,'CheckpointPath',tempdir)

%第2步，Training a Fast R-CNN Network using the RPN from step1

trainingOptions('sgdm','Maxepochs',10,'InitialLearnRate',1e-5,'CheckpointPath',tempdir)

%第3步，Re-training RPN using weight sharing with Fast R-CNN

trainingOptions('sgdm','Maxepochs',10,'InitialLearnRate',1e-6,'CheckpointPath',tempdir)

%第4步，Re-training Fast R-CNN using updated RPN

trainingOptions('sgdm','Maxepochs',10,'InitialLearnRate',1e-6,'CheckpointPath',tempdir)

];

%设置模型的本地存储

doTrainingAndEval = 1;

if doTrainingAndEval

%训练R-CNN神经网络，神经网络工具箱提供了3个函数

%(1)trainRCNNObjectDetector,训练快且检测满，允许指定proposaFcn

%(2)trainFastRCNNObjectDetector,速度较快，允许指定proposaFcn

%(3)trainFasterRCNNObjectDetector,优化运行性能，不需要指定proposaFcn

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