opencv3.3调用 caffe mnist 模型进行手写数字的分类

上一篇博文，正确跑通了mnist数据集，得到了lenet_iter_5000.caffemodel  lenet_iter_5000.solverstate  lenet_iter_10000.caffemodel  lenet_iter_10000.solverstate 四个文件

一、按照下面步骤修改caffe-master\examples\mnist文件夹中的  lenet_train_test.prototxt 文件

1、去掉数据输入层，即将top 为 “data” 的layers 去掉。 即将下面的内容删掉   

layer {  
  name: "mnist"  
  type: "Data"  
  top: "data"  
  top: "label"  
  include {  
    phase: TRAIN  
  }  
  transform_param {  
    scale: 0.00390625  
  }  
  data_param {  
    source: "examples/mnist/mnist_train_lmdb"  
    batch_size: 64  
    backend: LMDB  
  }  
}  
layer {  
  name: "mnist"  
  type: "Data"  
  top: "data"  
  top: "label"  
  include {  
    phase: TEST  
  }  
  transform_param {  
    scale: 0.00390625  
  }  
  data_param {  
    source: "examples/mnist/mnist_test_lmdb"  
    batch_size: 100  
    backend: LMDB  
  }  
}  

2、重新建立输入，即添加下面的内容

input: "data"  
input_shape {  
  dim: 1   # batchsize  
  dim: 1   # number of channels   
  dim: 28  # width  
  dim: 28  # height  
} 

3、去掉输出层，即将bottom 为 “label” 的layers 去掉，即将下面的内容删掉

layer {  
  name: "accuracy"  
  type: "Accuracy"  
  bottom: "ip2"  
  bottom: "label"  
  top: "accuracy"  
  include {  
    phase: TEST  
  }  
}  
layer {  
  name: "loss"  
  type: "SoftmaxWithLoss"  
  bottom: "ip2"  
  bottom: "label"  
  top: "loss"  
} 

4、重新建立输出，即添加下面的内容

layer {  
  name: "prob"  
  type: "Softmax"  
  bottom: "ip2"  
  top: "prob"  
}  

二、使用cv::dnn 里的API加载model, 输入图片，进行测试

#include <iostream>  
#include <opencv2/opencv.hpp>  
#include <opencv2/dnn.hpp>  

using namespace std;
using namespace cv;
using namespace cv::dnn;

/* Find best class for the blob (i. e. class with maximal probability) */
static void getMaxClass(const Mat &probBlob, int *classId, double *classProb)
{
	Mat probMat = probBlob.reshape(1, 1);
	Point classNumber;
	minMaxLoc(probMat, NULL, classProb, NULL, &classNumber);
	*classId = classNumber.x;
}

int main()
{
	string modelTxt = "lenet_train_test.prototxt";
	string modelBin = "lenet_iter_10000.caffemodel";
	//read image  
	Mat img = imread("4.bmp", 0);

	Mat inputBlob = dnn::blobFromImage(img, 0.00390625f, Size(28, 28), Scalar(), false); //Convert Mat to batch of images
	dnn::Net net;
	try{
		net = dnn::readNetFromCaffe(modelTxt, modelBin);
	}
	catch (cv::Exception &ee){
		cerr << "Exception: " << ee.what() << endl;
		if (net.empty()){
			cout << "Can't load the network by using the flowing files:" << endl;
			cout << "modelTxt: " << modelTxt << endl;
			cout << "modelBin: " << modelBin << endl; exit(-1);
		}
	}
	Mat pred;
	net.setInput(inputBlob, "data");//set the network input, "data" is the name of the input layer   
	pred = net.forward("prob");//compute output, "prob" is the name of the output layer   
	cout << pred << endl; int classId; double classProb; getMaxClass(pred, &classId, &classProb);
	cout << "Best Class: " << classId << endl;
	cout << "Probability: " << classProb * 100 << "%" << endl;
}

三、参考博文

http://blog.csdn.net/sushiqian/article/details/78555891
https://www.cnblogs.com/messier/p/7997951.html#_caption_1

https://docs.opencv.org/master/d5/de7/tutorial_dnn_googlenet.html

工程代码及caffemodel的下载