function [bboxes,scores,labelsIndex] = yolov3Detect(matFile, im，…networkInputSize、networkOutputs confidenceThreshold,……yolov3Detect函数检测图像中的边界框、分数和% labelsIndex。Preprocess Data该示例将所有预处理转换应用到训练期间应用的数据集%，数据增强除外。由于示例%使用了预先训练的YOLO v3网络，因此输入数据必须是原始数据的代表性%，并且不进行修改，以便进行无偏评估。具体来说，下面的预处理操作被应用到%输入数据。% 1。将图像调整为网络输入的大小，因为图像比networkInputSize大%。2.在%[0 1]范围内缩放图像像素。 3. Convert the resized and rescaled image to a dlarray object. im = dlarray(preprocessData(im, networkInputSize), "SSCB"); imageSize = size(im,[1,2]); %% Define Anchor Boxes % Specify the anchor boxes estimated on the basis of the preprocessed % training data used when training the YOLO v3 network. These anchor box % values are same as mentioned in "Object Detection Using YOLO v3 Deep % Learning" example. For details on estimating anchor boxes, see "Anchor % Boxes for Object Detection". anchors = [ 41 34; 163 130; 98 93; 144 125; 33 24; 69 66]; % Specify anchorBoxMasks to select anchor boxes to use in both the % detection heads of the YOLO v3 network. anchorBoxMasks is a cell array of % size M-by-1, where M denotes the number of detection heads. Each % detection head consists of a 1-by-N array of row index of anchors in % anchorBoxes, where N is the number of anchor boxes to use. Select anchor % boxes for each detection head based on size-use larger anchor boxes at % lower scale and smaller anchor boxes at higher scale. To do so, sort the % anchor boxes with the larger anchor boxes first and assign the first % three to the first detection head and the next three to the second % detection head. area = anchors(:, 1).*anchors(:, 2); [~, idx] = sort(area, 'descend'); anchors = anchors(idx, :); anchorBoxMasks = {[1,2,3],[4,5,6]}; %% Predict on Yolov3 % Predict and filter the detections based on confidence threshold. predictions = yolov3Predict(matFile,im,networkOutputs,anchorBoxMasks); %% Generate Detections % indices corresponding to x,y,w,h predictions for bounding boxes anchorIndex = 2:5; tiledAnchors = generateTiledAnchors(predictions,anchors,anchorBoxMasks,... anchorIndex); predictions = applyAnchorBoxOffsets(tiledAnchors, predictions,... networkInputSize, anchorIndex); [bboxes,scores,labelsIndex] = generateYOLOv3DetectionsForCodegen(predictions,... confidenceThreshold, overlapThreshold, imageSize, classes); end function YPredCell = yolov3Predict(matFile,im,networkOutputs,anchorBoxMask) % Predict the output of network and extract the confidence, x, y, % width, height, and class. % load the deep learning network for prediction persistent net; if isempty(net) net = coder.loadDeepLearningNetwork(matFile); end YPredictions = cell(coder.const(networkOutputs), 1); [YPredictions{:}] = predict(net, im); YPredCell = extractPredictions(YPredictions, anchorBoxMask); % Apply activation to the predicted cell array. YPredCell = applyActivations(YPredCell); end

代码生成成功:查看报告

function YPredCell = applyActivations(YPredCell) %#codegen % Copyright 2020-2021 The MathWorks, Inc. numCells = size(YPredCell, 1);for iCell = 1:numCells for idx = 1:3 YPredCell{iCell,idx} = sigmoidActivation(YPredCell{iCell,idx});end end for iCell = 1:numCells for idx = 4:5 YPredCell{iCell, idx} = exp(YPredCell{iCell, idx});end end for iCell = 1:numCells YPredCell{iCell, 6} = sigmoidActivation(YPredCell{iCell, 6});end end function out = sigmoidActivation(x) out = 1 /(1+exp(-x));结束

function predictions = extractPredictions(YPredictions, anchorBoxMask) %#codegen % Copyright 2020-2021 The MathWorks, Inc. numPredictionHeads = size(YPredictions, 1);预测=细胞(numPredictionHeads 6);for ii = 1:numPredictionHeads %获取特性大小所需的信息。numChannelsPred =大小(YPredictions {2}, 3);numAnchors =大小(anchorBoxMask {2}, 2);numPredElemsPerAnchors = numChannelsPred / numAnchors;allIds = (1: numChannelsPred);步= numPredElemsPerAnchors;endIdx = numChannelsPred;YPredictionsData = extractdata (YPredictions {2}); % X positions. startIdx = 1; predictions{ii,2} = YPredictionsData(:,:,startIdx:stride:endIdx,:); xIds = startIdx:stride:endIdx; % Y positions. startIdx = 2; predictions{ii,3} = YPredictionsData(:,:,startIdx:stride:endIdx,:); yIds = startIdx:stride:endIdx; % Width. startIdx = 3; predictions{ii,4} = YPredictionsData(:,:,startIdx:stride:endIdx,:); wIds = startIdx:stride:endIdx; % Height. startIdx = 4; predictions{ii,5} = YPredictionsData(:,:,startIdx:stride:endIdx,:); hIds = startIdx:stride:endIdx; % Confidence scores. startIdx = 5; predictions{ii,1} = YPredictionsData(:,:,startIdx:stride:endIdx,:); confIds = startIdx:stride:endIdx; % Accumulate all the non-class indexes nonClassIds = [xIds yIds wIds hIds confIds]; % Class probabilities. % Get the indexes which do not belong to the nonClassIds classIdx = setdiff(allIds, nonClassIds, 'stable'); predictions{ii,6} = YPredictionsData(:,:,classIdx,:); end end

函数tiledAnchors = generateTiledAnchors(YPredCell，锚盒，…%生成平铺锚偏移量，用于将预测从YOLO % v3网格单元坐标转换为边界框坐标%#codegen % Copyright 2020-2021 the MathWorks, Inc. numPredictionHeads = size(YPredCell,1);tiledAnchors = cell(numPredictionHeads, size(anchorIndex, 2));for i = 1:numPredictionHeads anchors =主播盒(主播盒掩码{i}，:);[h, w ~ n] =大小(YPredCell{我1});[tiledAnchors{2},我tiledAnchors{1}我]= ndgrid (h - 0: 0: w1,…1:尺寸(锚,1),1:n);[~, ~, tiledAnchors{我3}]= ndgrid (h - 0: 0: w1,锚(:,2),1:n);[~, ~, tiledAnchors{我4}]= ndgrid (h - 0: 0: w1,锚(:1),1:n);结束结束

function YPredCell = applyanchorboxoffset (tiledAnchors,YPredCell，…inputImageSize,anchorIndex) %#codegen % Convert the predictions from the YOLO v3 grid cell coordinates to % bounding box coordinates % Copyright 2020-2021 the MathWorks, Inc. for i = 1:size(YPredCell,1) [h,w，~，] = size(YPredCell{i,1});YPredCell{我anchorIndex (1)} = (tiledAnchors{1}我+……YPredCell{我anchorIndex(1)})。/ w;YPredCell{我anchorIndex (2)} = (tiledAnchors{我2}+……YPredCell{我anchorIndex (2)}) / h;YPredCell{我anchorIndex (3)} = (tiledAnchors{3}我。*……YPredCell{我anchorIndex(3)})。/ inputImageSize (2);YPredCell{我anchorIndex (4)} = (tiledAnchors{4}我。*……YPredCell{我anchorIndex(4)})。/ inputImageSize (1); end end

函数图像=预处理数据（图像，targetSize）%s调整图像大小，并将像素缩放到0到1之间。%#codegen%版权所有2020-2021 The MathWorks，Inc.imgSize=尺寸（图像）；%将具有单通道的输入图像转换为3个通道。如果numel（imgSize）<1图像=repmat（图像，1,1,3）；结束图像=im2single（重新缩放（图像））；image=iLetterBoxImage（图像，编码器常数（targetSize（1:2））；结束函数Inew=iLetterBoxImage（I，targetSize）%LetterBoxImage通过保留输入图像I的宽度和高度%纵横比来返回调整大小的图像。“targetSize”是由%目标维度组成的1×2向量。%%输入I可以是uint8、uint16、int16、double、single或logical，并且必须是实的和非稀疏的。[Irow，Icol，Ichannels]=尺寸（I）；%计算纵横比。arI=Irow./Icol；%根据纵横比保留最大尺寸。如果arI<1 IcolFin=targetSize（1,2）；IrowFin=地板（IcolFin.*arI）；else IrowFin=目标尺寸（1,1）；IcolFin=地板（IrowFin./arI）；结束%调整输入图像的大小。Itmp=imresize（I[IrowFin，IcolFin]）；%使用灰色值初始化Inew。Inew=一（[targetSize，Ichannels]，'like'，I）。*0.5；%计算偏移量。如果arI<1 buff=targetSize（1,1）-IrowFin；else buff=目标尺寸（1,2）-IcolFin；结束%将调整大小的图像放置在画布图像上。如果（buff==0）Inew=Itmp；else buffVal=地板（buff/2）；如果arI<1 Inew（buffVal:buffVal+IrowFin-1，：，：）=Itmp；else Inew（：，buffVal:buffVal+IcolFin-1，：）=Itmp；结束