注意:该版本为CPU版本。
用到的caffe-windows来自:https://github.com/happynear/caffe-windows
先下载caffe-windows,解压;然后下载第三方库:https://pan.baidu.com/s/1eStyfrc 解压到caffe-windows-master,看起来是这样:caffe-windows-master\3rdparty
把3rdparty的bin加入环境变量或者复制里面的dll到build_cpu_only\caffelib下(cudnn的不需要)。
打开caffe-windows-master\src\caffe\proto,双击extract_proto.bat,然后用VS2013打开./build_cpu_only/MainBuilder.sln。请确保为Release x64
1.右键caffelib项目,重命名为:multi_recognition_cpu(按个人爱好,其他名字也行,不改也可以);再右键该项目——>属性——>配置属性——>常规:
配置类型修改为动态库(.dll),目标扩展名修改为.dll。
2.C/C++——>常规:
附加包含目录:
../../3rdparty/include
../../src
../../include
C/C++——>预处理器:
添加 MULTI_RECOGNITION_API_EXPORTS
3.链接器——>常规:
附加库目录:
../../3rdparty/lib
链接器——>输入:
去掉cuda和cudnn的lib(cu开头和cudnn开头的lib)
4.修改net.hpp和net.cpp
为了支持模型多输出,要知道输出的顺序,所以把输出blob的名字输出到控制台,打开net.hpp,给Net类添加:
protected:
std::vector<std::string> outputblobnames;
以及:
public:
inline std::vector<std::string> output_blobs_names() const
{
return outputblobnames;
}
net.cpp修改:(最后一行,把输出blob名字保存到vector中)
for (set<string>::iterator it = available_blobs.begin();
it != available_blobs.end(); ++it) {
LOG_IF(INFO, Caffe::root_solver())
<< "This network produces output " << *it;
net_output_blobs_.push_back(blobs_[blob_name_to_idx[*it]].get());
net_output_blob_indices_.push_back(blob_name_to_idx[*it]);
outputblobnames.push_back(*it);
}
这样,属性就配置好了代码也修改完了,再右键该项目,添加新建项,有四个:
classification.h
classification.cpp
multi_recognition_cpu.h
multi_recognition_cpu.cpp
classification.h:
#ifndef CLASSIFICATION_H_
#define CLASSIFICATION_H_
#include <caffe/caffe.hpp>
#include <opencv2/core/core.hpp>
#include <opencv2/highgui/highgui.hpp>
#include <opencv2/imgproc/imgproc.hpp>
#include <iosfwd>
#include <memory>
#include <utility>
#include <vector>
#include <iostream>
#include <string>
#include <time.h>
using namespace caffe;
using std::string;
typedef std::pair<int, float> Prediction;
class ClassifierImpl {
public:
ClassifierImpl(const string& model_file,
const string& trained_file,
const string& mean_file
);
std::vector<std::vector<Prediction> > Classify(const cv::Mat& img, int N = 2);
private:
void SetMean(const string& mean_file);
std::vector<std::vector<float> > Predict(const cv::Mat& img);
void WrapInputLayer(std::vector<cv::Mat>* input_channels);
void Preprocess(const cv::Mat& img,
std::vector<cv::Mat>* input_channels);
private:
shared_ptr<Net<float> > net_;
cv::Size input_geometry_;
int num_channels_;
cv::Mat mean_;
};
#endif
classification.cpp:
#include "classification.h"
ClassifierImpl::ClassifierImpl(const string& model_file,
const string& trained_file,
const string& mean_file) {
#ifdef CPU_ONLY
Caffe::set_mode(Caffe::CPU);
#else
Caffe::set_mode(Caffe::GPU);
#endif
/* Load the network. */
net_.reset(new Net<float>(model_file, TEST));
net_->CopyTrainedLayersFrom(trained_file);
CHECK_EQ(net_->num_inputs(), 1) << "Network should have exactly one input.";
std::cout << "Network have " << net_->num_outputs() << " outputs.\n";
vector<string> names = net_->output_blobs_names();
for (int n = 0; n < net_->num_outputs(); ++n)
{
std::cout << "Output " << n + 1 << ":" << names[n] << "; have " << net_->output_blobs()[n]->channels() << " outputs.\n";
}
Blob<float>* input_layer = net_->input_blobs()[0];
std::cout << "Input width:" << input_layer->width() << ";" << "Input height:" << input_layer->height() << "\n";
num_channels_ = input_layer->channels();
CHECK(num_channels_ == 3 || num_channels_ == 1)
<< "Input layer should have 1 or 3 channels.";
input_geometry_ = cv::Size(input_layer->width(), input_layer->height());
/* Load the binaryproto mean file. */
SetMean(mean_file);
}
static bool PairCompare(const std::pair<float, int>& lhs,
const std::pair<float, int>& rhs) {
return lhs.first > rhs.first;
}
/* Return the indices of the top N values of vector v. */
static std::vector<int> Argmax(const std::vector<float>& v, int N) {
std::vector<std::pair<float, int> > pairs;
for (size_t i = 0; i < v.size(); ++i)
pairs.push_back(std::make_pair(v[i], i));
std::partial_sort(pairs.begin(), pairs.begin() + N, pairs.end(), PairCompare);
std::vector<int> result;
for (int i = 0; i < N; ++i)
result.push_back(pairs[i].second);
return result;
}
/* Return the top N predictions. */
std::vector<std::vector<Prediction> > ClassifierImpl::Classify(const cv::Mat& img, int N) {
std::vector<std::vector<Prediction> > outputPredict;
std::vector<std::vector<float> > output = Predict(img);
for (auto bg = output.begin(); bg != output.end(); ++bg)
{
std::vector<int> maxN = Argmax(*bg, N);
std::vector<Prediction> predictions;
for (int i = 0; i < N; ++i) {
int idx = maxN[i];
predictions.push_back(std::make_pair(idx, (*bg)[idx]));
}
outputPredict.push_back(predictions);
predictions.clear();
maxN.clear();
}
return outputPredict;
}
/* Load the mean file in binaryproto format. */
void ClassifierImpl::SetMean(const string& mean_file) {
BlobProto blob_proto;
ReadProtoFromBinaryFileOrDie(mean_file.c_str(), &blob_proto);
Blob<float> mean_blob;
mean_blob.FromProto(blob_proto);
CHECK_EQ(mean_blob.channels(), num_channels_)
<< "Number of channels of mean file doesn't match input layer.";
std::vector<cv::Mat> channels;
float* data = mean_blob.mutable_cpu_data();
for (int i = 0; i < num_channels_; ++i) {
cv::Mat channel(mean_blob.height(), mean_blob.width(), CV_32FC1, data);
channels.push_back(channel);
data += mean_blob.height() * mean_blob.width();
}
cv::Mat mean;
cv::merge(channels, mean);
cv::Scalar channel_mean = cv::mean(mean);
mean_ = cv::Mat(input_geometry_, mean.type(), channel_mean);
}
std::vector<std::vector<float> > ClassifierImpl::Predict(const cv::Mat& img) {
Blob<float>* input_layer = net_->input_blobs()[0];
input_layer->Reshape(1, num_channels_,
input_geometry_.height, input_geometry_.width);
net_->Reshape();
std::vector<cv::Mat> input_channels;
WrapInputLayer(&input_channels);
Preprocess(img, &input_channels);
net_->ForwardPrefilled();
std::vector<std::vector<float> > outPredict;
for (int i = 0; i < net_->output_blobs().size(); ++i)
{
Blob<float>* output_layer = net_->output_blobs()[i];
const float* begin = output_layer->cpu_data();
const float* end = begin + output_layer->channels();
std::vector<float> temp(begin, end);
outPredict.push_back(temp);
temp.clear();
}
return outPredict;
}
void ClassifierImpl::WrapInputLayer(std::vector<cv::Mat>* input_channels) {
Blob<float>* input_layer = net_->input_blobs()[0];
int width = input_layer->width();
int height = input_layer->height();
float* input_data = input_layer->mutable_cpu_data();
for (int i = 0; i < input_layer->channels(); ++i) {
cv::Mat channel(height, width, CV_32FC1, input_data);
input_channels->push_back(channel);
input_data += width * height;
}
}
void ClassifierImpl::Preprocess(const cv::Mat& img,
std::vector<cv::Mat>* input_channels) {
cv::Mat sample;
if (img.channels() == 3 && num_channels_ == 1)
cv::cvtColor(img, sample, CV_BGR2GRAY);
else if (img.channels() == 4 && num_channels_ == 1)
cv::cvtColor(img, sample, CV_BGRA2GRAY);
else if (img.channels() == 4 && num_channels_ == 3)
cv::cvtColor(img, sample, CV_BGRA2BGR);
else if (img.channels() == 1 && num_channels_ == 3)
cv::cvtColor(img, sample, CV_GRAY2BGR);
else
sample = img;
cv::Mat sample_resized;
if (sample.size() != input_geometry_)
cv::resize(sample, sample_resized, input_geometry_);
else
sample_resized = sample;
cv::Mat sample_float;
if (num_channels_ == 3)
sample_resized.convertTo(sample_float, CV_32FC3);
else
sample_resized.convertTo(sample_float, CV_32FC1);
cv::Mat sample_normalized;
cv::subtract(sample_float, mean_, sample_normalized);
cv::split(sample_normalized, *input_channels);
CHECK(reinterpret_cast<float*>(input_channels->at(0).data)
== net_->input_blobs()[0]->cpu_data())
<< "Input channels are not wrapping the input layer of the network.";
}
导出类:
multi_recognition_cpu.h:
#ifndef MULTI_RECOGNITION_CPU_H_
#define MULTI_RECOGNITION_CPU_H_
#ifdef MULTI_RECOGNITION_API_EXPORTS
#define MULTI_RECOGNITION_API __declspec(dllexport)
#else
#define MULTI_RECOGNITION_API __declspec(dllimport)
#endif
#include <opencv2/core/core.hpp>
#include <opencv2/highgui/highgui.hpp>
#include <opencv2/imgproc/imgproc.hpp>
#include <string>
#include <vector>
#include <iostream>
#include <io.h>
class ClassifierImpl;
using std::string;
using std::vector;
typedef std::pair<int, float> Prediction;
class MULTI_RECOGNITION_API MultiClassifier
{
public:
MultiClassifier(const string& model_file,
const string& trained_file,
const string& mean_file);
~MultiClassifier();
std::vector<std::vector<Prediction> >Classify(const cv::Mat& img, int N = 2);
void getFiles(std::string path, std::vector<std::string>& files);
private:
ClassifierImpl *Impl;
};
#endif
multi_recognition_cpu.cpp:
#include "multi_recognition_cpu.h"
#include "classification.h"
MultiClassifier::MultiClassifier(const string& model_file, const string& trained_file, const string& mean_file)
{
Impl = new ClassifierImpl(model_file, trained_file, mean_file);
}
MultiClassifier::~MultiClassifier()
{
delete Impl;
}
std::vector<std::vector<Prediction> > MultiClassifier::Classify(const cv::Mat& img, int N /* = 2 */)
{
return Impl->Classify(img, N);
}
void MultiClassifier::getFiles(string path, vector<string>& files)
{
//文件句柄
long hFile = 0;
//文件信息
struct _finddata_t fileinfo;
string p;
if ((hFile = _findfirst(p.assign(path).append("\\*").c_str(), &fileinfo)) != -1)
{
do
{
if ((fileinfo.attrib & _A_SUBDIR))
{
if (strcmp(fileinfo.name, ".") != 0 && strcmp(fileinfo.name, "..") != 0)
getFiles(p.assign(path).append("\\").append(fileinfo.name), files);
}
else
{
files.push_back(p.assign(path).append("\\").append(fileinfo.name));
}
} while (_findnext(hFile, &fileinfo) == 0);
_findclose(hFile);
}
}
右键项目,生成就可以了。
最后得到:
模型可以有多个输出:
封装的代码下载地址:caffe-windows-cpu