Darknet框架简介

怼烎@ 2022-10-29 15:20 539阅读 0赞

### Darknet框架简介 ###

*  前言：为什么选择darknet？
 *   *  linux端
 *  一、下载源码及权重文件
 *  二、测试
 *  三、训练自己的数据集
 *   *  windows端
     *  编译darknet：
     *  main.c
     *   *   *  第二版
     *  
     *  封装及调用——方便工程化。

darknet是一个较为轻型的完全基于C与CUDA的开源深度学习框架，其主要特点就是容易安装，没有任何依赖项（OpenCV都可以不用），移植性非常好，支持CPU与GPU两种计算方式。

# 前言：为什么选择darknet？ #

*  相比于TensorFlow来说，darknet并没有那么强大，但这也成了darknet的优势：
 *  darknet完全由C语言实现，没有任何依赖项，当然可以使用OpenCV，但只是用其来显示图片、为了更好的可视化；
 *  darknet支持CPU（所以没有GPU也不用紧的）与GPU（CUDA/cuDNN，使用GPU当然更块更好了）；
 *  正是因为其较为轻型，没有像TensorFlow那般强大的API，所以给我的感觉就是有另一种味道的灵活性，适合用来研究底层，可以更为方便的从底层对其进行改进与扩展；
 *  darknet的实现与caffe的实现存在相似的地方，熟悉了darknet，相信对上手caffe有帮助

--------------------

## linux端 ##

# 一、下载源码及权重文件 #

官方darknet源码下载网址：[https://github.com/pjreddie/darknet][https_github.com_pjreddie_darknet]  
官方权重下载地址：[https://pjreddie.com/media/files/yolov3\_tiny.weights ][https_pjreddie.com_media_files_yolov3_tiny.weights]  
csdn下载地址：[yolov3-tiny 权重文件][yolov3-tiny]

主要参考：[详细步骤！使用yolov3-tiny训练，测试、验证VOC数据集][yolov3-tiny_VOC]

https://blog.csdn.net/xiao\_lxl/article/details/85047103

下载后，首先查看makefile  
![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L21hb19odWlfZmVp_size_16_color_FFFFFF_t_70]  
可以根据自己实际情况进行更改，用下面命令进行编译

make clean;make all

编译完成后生成darknet 可执行文件及动态库。

如下：  
![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L21hb19odWlfZmVp_size_16_color_FFFFFF_t_70 1]  
文件夹说明

*  1.cfg文件夹内是一些模型的架构，每个cfg文件类似与caffe的prototxt文件，通过该文件定义的整个模型的架构
 *  2.data文件夹内放置了一些label文件，如coco9k的类别名等，和一些样例图（该文件夹主要为演示用，或者是直接训练coco等对应数据集时有用，如果要用自己的数据自行训练，该文件夹内的东西都不是我们需要的）
 *  3.src文件夹内全是最底层的框架定义文件，所有层的定义等最基本的函数全部在该文件夹内，可以理解为该文件夹就是框架的源码；
 *  4.examples文件夹是更为高层的一些函数，如检测函数，识别函数等，这些函数直接调用了底层的函数，我们经常使用的就是example中的函数；
 *  5.include文件夹，顾名思义，存放头文件的地方；
 *  6.python文件夹里是使用python对模型的调用方法，基本都在darknet.py中。当然，要实现python的调用，还需要用到darknet的动态库libdarknet.so，这个动态库稍后再介绍；
 *  7.scripts文件夹中是一些脚本，如下载coco数据集，将voc格式的数据集转换为训练所需格式的脚本等。
 *  8.除了license文件，剩下的就是Makefile文件，如下图，在问价开头有一些选项，把你需要使用的选项设为1即可。

# 二、测试 #

./darknet detector test cfg/coco.data cfg/yolov3-tiny.cfg weights/yolov3-tiny.weights  data/dog.jpg

![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L21hb19odWlfZmVp_size_16_color_FFFFFF_t_70 2]  
出现上述的结果，说明运行正常。

# 三、训练自己的数据集 #

待续。。。

--------------------

## windows端 ##

参考：

## 编译darknet： ##

[https://blog.csdn.net/qq\_28663849/article/details/107362445][https_blog.csdn.net_qq_28663849_article_details_107362445]

CPU版本  
首先打开makefile中，特别是OPENMP和AVX，速度快

OPENCV=1
    AVX=1
    OPENMP=1

*  1、先运行 编译darknet.sln，生成解决方案，**release版本**，生成darknet.exe。
 *  2、基于上一版本运行 编译yolo\_cpp\_dll\_no\_gpu.sln，生成解决方案，**release版本**，生成yolo\_cpp\_dll.lib和yolo\_cpp\_dll.dll。（注意是yolo\_cpp\_dll）  
    3、复制生成的yolo\_cpp\_dll.lib和yolo\_cpp\_dll.dll到新工程。  
    ![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L21hb19odWlfZmVp_size_16_color_FFFFFF_t_70 3]  
    1、动态链接库（均在darknet-master\\build\\darknet\\x64目录下）  
    (1)yolo\_cpp\_dll.lib  
    (2)yolo\_cpp\_dll.dll  
    (3)pthreadGC2.dll  
    (4)pthreadVC2.dll  
    2、OpenCV库（取决于使用debug还是release模式）——或者自己直接在属性管理器中配置。  
    (1)opencv\_world340d.dll  
    (2)opencv\_world340.dll  
    3、YOLO模型文件（第一个文件在darknet-master\\build\\darknet\\x64\\data目录下，第二个文件在darknet-master\\build\\darknet\\x64目录下，第三个文件需要自己下载，下载链接见前一篇文章）  
    (1)coco.names  
    (2)yolov3.cfg  
    (3)yolov3.weights  
    4、头文件  
    (1)yolo\_v2\_class.hpp  
    头文件包含了动态链接库中具体的类的定义，调用时需要引用，这个文件在darknet-master\\include目录下的yolo\_v2\_class.hpp中，将其复制到记事本保存成.hpp文件即可。

**参考：**  
原文链接：[https://blog.csdn.net/stjuliet/article/details/87884976][https_blog.csdn.net_stjuliet_article_details_87884976]

## main.c ##

#include <iostream>
    
    
    #define OPENCV
    //#define GPU
    
    
    #include "yolo_v2_class.hpp" //引用动态链接库中的头文件
    #include <opencv2/opencv.hpp>
    #include "opencv2/highgui/highgui.hpp"
    
    //#pragma comment(lib, "opencv_world340d.lib") //引入OpenCV链接库
    //#pragma comment(lib, "yolo_cpp_dll_no_gpu.lib") //引入YOLO动态链接库
    #pragma comment(lib, "yolo_cpp_dll.lib") //引入YOLO动态链接库
    //以下两段代码来自yolo_console_dll.sln
    void draw_boxes(cv::Mat mat_img, std::vector<bbox_t> result_vec, std::vector<std::string> obj_names,
    	int current_det_fps , int current_cap_fps )
    { 
    	int const colors[6][3] = {  {  1,0,1 },{  0,0,1 },{  0,1,1 },{  0,1,0 },{  1,1,0 },{  1,0,0 } };
    
    	for (auto &i : result_vec) 
    	{ 
    		cv::Scalar color = obj_id_to_color(i.obj_id);
    		cv::rectangle(mat_img, cv::Rect(i.x, i.y, i.w, i.h), color, 2);
    		if (obj_names.size() > i.obj_id) 
    		{ 
    			std::string obj_name = obj_names[i.obj_id];
    			if (i.track_id > 0) obj_name += " - " + std::to_string(i.track_id);
    			cv::Size const text_size = getTextSize(obj_name, cv::FONT_HERSHEY_COMPLEX_SMALL, 1.2, 2, 0);
    			int const max_width = (text_size.width > i.w + 2) ? text_size.width : (i.w + 2);
    			cv::rectangle(mat_img, cv::Point2f(std::max((int)i.x - 1, 0), std::max((int)i.y - 30, 0)),
    				cv::Point2f(std::min((int)i.x + max_width, mat_img.cols - 1), std::min((int)i.y, mat_img.rows - 1)),
    				color, CV_FILLED, 8, 0);
    			putText(mat_img, obj_name, cv::Point2f(i.x, i.y - 10), cv::FONT_HERSHEY_COMPLEX_SMALL, 1.2, cv::Scalar(0, 0, 0), 2);
    		}
    	}
    	if (current_det_fps >= 0) 
    	{ 
    		std::string fps_str = "FPS detection: " + std::to_string(current_det_fps) ;
    		putText(mat_img, fps_str, cv::Point2f(10, 20), cv::FONT_HERSHEY_COMPLEX_SMALL, 1.2, cv::Scalar(50, 255, 0), 2);
    	}
    	if (current_cap_fps >= 0)
    	{ 
    		std::string fps_str = "FPS capture: " + std::to_string(current_cap_fps);
    		putText(mat_img, fps_str, cv::Point2f(10, 50), cv::FONT_HERSHEY_COMPLEX_SMALL, 1.2, cv::Scalar(50, 255, 0), 2);
    	}
    }
    
    std::vector<std::string> objects_names_from_file(std::string const filename) 
    { 
    	std::ifstream file(filename);
    	std::vector<std::string> file_lines;
    	if (!file.is_open()) return file_lines;
    	for (std::string line; getline(file, line);) file_lines.push_back(line);
    	std::cout << "object names loaded \n";
    	return file_lines;
    }
    
    int main()
    { 
    	std::string names_file = "coco.names";
    	std::string cfg_file = "yolov3-tiny.cfg";
    	std::string weights_file = "yolov3-tiny.weights";
    	Detector detector(cfg_file, weights_file, 0); //初始化检测器
    //std::vector<std::string> obj_names = objects_names_from_file(names_file); //调用获得分类对象名称
    //或者使用以下四行代码也可实现读入分类对象文件
    	std::vector<std::string> obj_names;
    	std::ifstream ifs(names_file.c_str());
    	std::string line;
    	while (getline(ifs, line)) obj_names.push_back(line);
    	//测试是否成功读入分类对象文件
    	for (size_t i = 0; i < obj_names.size(); i++)
    	{ 
    		std::cout << obj_names[i] << std::endl;
    	}
    
    	cv::VideoCapture capture;
    	capture.open("bsd.mp4");
    	if (!capture.isOpened())
    	{ 
    		printf("文件打开失败");
    	}
    	cv::Mat frame;
    	float fps = 0;
    	while (true)
    	{ 
    		int begin, duration;
    		begin = clock();
    		capture >> frame;
    		//if (frame.empty() == NULL)
    		//{ 
    		// printf("Image is empty");
    		// return 0;
    		//}
    
    
    		duration = clock() - begin;
    		//printf("读图用时约: %d毫秒 \n", duration * 1000 / CLOCKS_PER_SEC);
    		//cv::resize(frame, frame, cv::Size(640,360));
    		std::vector<bbox_t> result_vec = detector.detect(frame);
    		duration = clock() - begin;
    		printf("识别用时约: %d毫秒 \n", duration * 1000 / CLOCKS_PER_SEC);
    		fps =(float) 1000.0 / duration;
    		draw_boxes(frame, result_vec, obj_names,fps, fps);
    		//duration = clock() - begin;
    		//printf("画框用时约: %d毫秒 \n", duration * 1000 / CLOCKS_PER_SEC);
    
    		//cv::namedWindow("bsd", 0);
    		//cv::imshow("bsd", frame);
    		//cv::waitKey(1);
    		//duration = clock() - begin;
    		//printf("显示用时约: %d毫秒 \n", duration * 1000 / CLOCKS_PER_SEC);
    		printf("\n\n\n");
    
    	}
    	return 0;
    }

#### 第二版 ####

/***************************************************************************************** 1、盲区检测 2021.3.17： *****************************************************************************************/
    #define OPENCV/*必须放置在链接的头文件前面*/
    //#define GPU
    
    #include <iostream>
    #include <stdlib.h> //srand()和rand()函数 
    #include<windows.h>
    #include <fstream> 
    
    #include "yolo_v2_class.hpp" //引用动态链接库中的头文件
    
    /*opencv相关*/
    #include <opencv2/core/core.hpp> 
    #include <opencv2/opencv.hpp>
    #include "opencv2/highgui/highgui.hpp"
    #include <opencv2/imgproc/imgproc.hpp> 
    #include <opencv2/objdetect/objdetect.hpp> 
    #include <opencv2/ml/ml.hpp> 
    #include <opencv2\imgproc\types_c.h>
    
    
    using namespace cv;
    using namespace std;
    /**************** 打印相关组件 start ! ********************************************************************/
    
    #define debug_show_picture 1 //是否显示部分调试图片，方便调试
    /*打印等级，修改后面的宏定义可以改变函数输出打印等级*/
    #define ALG_PRTINT_LEVER PRINT_LEVEL_UNLIMIT
    
    #define ALG_PRTINT(...) SAL_printf(__VA_ARGS__)
    #define ALG_PRT(...) ALG_PRTINT(__FUNCTION__, __LINE__, PRINT_LEVEL_UNLIMIT, __VA_ARGS__)
    #define ALG_DBG(...) ALG_PRTINT(__FUNCTION__, __LINE__, PRINT_LEVEL_DBG, __VA_ARGS__)
    #define ALG_WAR(...) ALG_PRTINT(__FUNCTION__, __LINE__, PRINT_LEVEL_WRN, __VA_ARGS__)
    #define ALG_ERR(...) ALG_PRTINT(__FUNCTION__, __LINE__, PRINT_LEVEL_ERR, __VA_ARGS__)
    
    /*********************************************************************************************** * @enum HAT_SAL_PRT_LEVEL_E * @brief 打印输出的等级 ***************************************************************************************************/
    typedef enum _PRT_LEVEL_E_
    { 
    	PRINT_LEVEL_ERR = 0,        /*错误打印信息*/
    	PRINT_LEVEL_WRN = 1,        /*警告打印信息*/
    	PRINT_LEVEL_DBG = 2,         /*调试打印信息*/
    	PRINT_LEVEL_UNLIMIT = 3,    /*无限制打印信息*/
    	PRINT_LEVEL_NOPRT = 4,      /*没有打印信息*/
    } PRT_LEVEL_E;
    
    /******************************************************************************* Function: SAL_printf Description: 该函数能够通过设置的打印等级ALG_PRTINT_LEVER，来控制是否输出相关语句 Input: Output: Return: 0: Successful ohters: Failed *******************************************************************************/
    void SAL_printf(const char *pFun, UINT line, PRT_LEVEL_E levelParam, const char *fmt, ...)
    { 
    	static INT8 g_printfInfo[4][16] = {  "ERR", "WAR", "DBG", "INF" };
    	va_list p;
    	if (ALG_PRTINT_LEVER == PRINT_LEVEL_NOPRT || levelParam == PRINT_LEVEL_NOPRT)
    	{ 
    		return;
    	}
    	if (levelParam <= ALG_PRTINT_LEVER)
    	{ 
    		va_start(p, fmt);
    		printf("[DIYUN_ALG][%s][%s][%4d] ", g_printfInfo[levelParam], pFun, line);
    		vprintf(fmt, p);
    		va_end(p);
    	}
    }
    /****************************** 打印相关组件 end *************************************************************************/
    
    
    
    
    
    
    
    
    
    //#pragma comment(lib, "opencv_world340d.lib") //引入OpenCV链接库
    //#pragma comment(lib, "yolo_cpp_dll_no_gpu.lib") //引入YOLO动态链接库
    #pragma comment(lib, "yolo_cpp_dll.lib") //引入YOLO动态链接库
    
    #ifdef max
    #undef max
    #endif
    
    #ifdef min
    #undef min
    #endif
    
    //以下两段代码来自yolo_console_dll.sln
    void draw_boxes(Mat mat_img, vector<bbox_t> result_vec, vector<string> obj_names,
    	int current_det_fps , int current_cap_fps )
    { 
    	int const colors[6][3] = {  {  1,0,1 },{  0,0,1 },{  0,1,1 },{  0,1,0 },{  1,1,0 },{  1,0,0 } };
    
    	for (auto &i : result_vec) 
    	{ 
    		Scalar color = obj_id_to_color(i.obj_id);
    		rectangle(mat_img, Rect(i.x, i.y, i.w, i.h), color, 2);
    		if (obj_names.size() > i.obj_id) 
    		{ 
    			string obj_name = obj_names[i.obj_id];
    			if (i.track_id > 0) obj_name += " - " + to_string(i.track_id);
    			Size const text_size = getTextSize(obj_name, FONT_HERSHEY_COMPLEX_SMALL, 1.2, 2, 0);
    			int const max_width = (text_size.width > i.w + 2) ? text_size.width : (i.w + 2);
    			rectangle(mat_img, Point2f(max((int)i.x - 1, 0), max((int)i.y - 30, 0)),
    				Point2f(min((int)i.x + max_width, mat_img.cols - 1), min((int)i.y, mat_img.rows - 1)),
    				color, CV_FILLED, 8, 0);
    			putText(mat_img, obj_name, Point2f(i.x, i.y - 10), FONT_HERSHEY_COMPLEX_SMALL, 1.2, Scalar(0, 0, 0), 2);
    		}
    	}
    	if (current_det_fps >= 0) 
    	{ 
    		string fps_str = "FPS detection: " + to_string(current_det_fps) ;
    		putText(mat_img, fps_str, Point2f(10, 20), FONT_HERSHEY_COMPLEX_SMALL, 1.2, Scalar(50, 255, 0), 2);
    	}
    	if (current_cap_fps >= 0)
    	{ 
    		string fps_str = "FPS capture: " + to_string(current_cap_fps);
    		putText(mat_img, fps_str, Point2f(10, 50), FONT_HERSHEY_COMPLEX_SMALL, 1.2, Scalar(50, 255, 0), 2);
    	}
    }
    
    vector<string> objects_names_from_file(string const filename) 
    { 
    	ifstream file(filename);
    	vector<string> file_lines;
    	if (!file.is_open()) return file_lines;
    	for (string line; getline(file, line);) file_lines.push_back(line);
    	cout << "object names loaded \n";
    	return file_lines;
    }
    
    //点(浮点型) 
    typedef struct _POINT_F_
    { 
    	float x;
    	float y;
    }POINT_F;
    
    //多边形(浮点型)
    typedef struct _VCA_POLYGON_F_
    { 
    	unsigned int  vertex_num;                  //顶点数
    	POINT_F   point[5];   //顶点
    }VCA_POLYGON_F;
    
    //前后视图片切换索引
    typedef enum _BSD_WARNNING_TYPE_
    { 
    	BSD_WARNNING_INVALID = 0,              // 无效值
    	BSD_WARNNING_WARN = 1,              // BSD提醒（超车）
    	BSD_WARNNING_ALARM = 2,              // BSD预警（无超车要求）
    	BSD_WARNNING_ICW = 3,              // 摄像头失效报警
    	BSD_EVENT_MOD = 4,              // 移动障碍物检测
    	BSD_EVENT_MOD_ALL = 5,              // 移动障碍物检测(全景)
    	BSD_EVENT_DOW = 6,              // 开门预警检测
    	BSD_WARNNING_END
    } BSD_WARNNING_TYPE;
    
    // BSDC报警区域枚举
    typedef enum _VCA_REAR_VIEW_STATE_
    { 
    	VCA_MOD_INVALID = 0,           // 无障碍物（无效）
    	VCA_MOD_OBS = 1,           // 近处有障碍物
    	VCA_MOD_FAR = 2,           // 远处有障碍物
    	VCA_MOD_LEFT = 3,           // 左侧有障碍物
    	VCA_MOD_RIGHT = 4,           // 右侧有障碍物
    	VCA_MOD_LEFT_LEVEL_TWO = 5,    // 左侧有障碍物二级报警
    	VCA_MOD_RIGHT_LEVEL_TWO = 6,   // 右侧有障碍物二级报警
    	VCA_MOD_LEVEL1 = 11,           // 一级区域——最危险
    	VCA_MOD_LEVEL2 = 12,           // 二级区域——中等危险
    	VCA_MOD_LEVEL3 = 13,           // 三级区域——较安全
    
    	VCA_MOD_END
    } VCA_REAR_VIEW_STATE;
    
    /* VCA_BSD配置参数 */
    typedef struct _BSD_REGION_PARAM_SINGLE_
    { 
    	BSD_WARNNING_TYPE     	event_type;/*报警类型*/
    	VCA_REAR_VIEW_STATE   	view_state;			/* 报警区域级别 */
    	VCA_POLYGON_F         	poly;      /*报警框顶点*/
    	short                 	sensity;
    	char                  	reserved[2];
    } BSD_REGION_PARAM_SINGLE;
    
    /* BSD配置参数 */
    typedef struct _BSD_PARAM_ALL_
    { 
    	int                     region_num;
    	BSD_REGION_PARAM_SINGLE	region[4];
    	int			time_type;/*0 1 2*/
    	char					mirror_flag;		/* 图像和结果是否需要镜像，1为镜像，0不镜像 */
    } VCA_BSD_PARAM_ALL;
    
    // 侧视、车尾的报警区域配置参数
    typedef struct _VCA_ADAS_BSDC_MOD_INPUT_PARAM_
    { 
    	int                   whole_w;
    	int                   whole_h;
    	int                   poly_valid[3];  // 报警区域是否有效标志位
    	VCA_POLYGON_F         bsd_poly[3];    // BSD报警区域坐标
    	VCA_REAR_VIEW_STATE   view_state[3];  // 报警区域级别
    } VCA_ADAS_BSDC_MOD_INPUT_PARAM;
    
    /*全局变量，用于bsd参数配置传递等*/
    VCA_ADAS_BSDC_MOD_INPUT_PARAM gBsd_init_para = {  0 };
    
    /******************************************************************************* 函数名称: BsdInitParam 描 述: 该函数初始化BSD报警参数配置，主要是报警框的设置 输入:输入宽高 输出: 返回值: 0: Successful ohters: Failed *******************************************************************************/
    int BsdInitParam(int  width,int height)
    { 
    	bool OK = 0;/*返回值初始化*/
    	int ori_w = width;
    	int ori_h = height;
    
    	VCA_ADAS_BSDC_MOD_INPUT_PARAM *pBsd_init_para = NULL;
    	pBsd_init_para = &gBsd_init_para;
    #if 0
    	/*左视*/
    	//一级区域(一般为最小、最靠近本车的区域)
    	pBsd_init_para->poly_valid[0] = 1;                     //该区域有效
    	pBsd_init_para->view_state[0] = VCA_MOD_LEVEL3;          //一级区域
    	pBsd_init_para->bsd_poly[0].vertex_num = 5;
    	pBsd_init_para->bsd_poly[0].point[0].x = (float)1100 / ori_w;
    	pBsd_init_para->bsd_poly[0].point[1].x = (float)1255 / ori_w;
    	pBsd_init_para->bsd_poly[0].point[2].x = (float)1083 / ori_w;
    	pBsd_init_para->bsd_poly[0].point[3].x = (float)643 / ori_w;
    	pBsd_init_para->bsd_poly[0].point[4].x = (float)643 / ori_w;
    
    	pBsd_init_para->bsd_poly[0].point[0].y = (float)238 / ori_h;
    	pBsd_init_para->bsd_poly[0].point[1].y = (float)218 / ori_h;
    	pBsd_init_para->bsd_poly[0].point[2].y = (float)713 / ori_h;
    	pBsd_init_para->bsd_poly[0].point[3].y = (float)716 / ori_h;
    	pBsd_init_para->bsd_poly[0].point[4].y = (float)626 / ori_h;
    
    	//二级区域 
    	pBsd_init_para->poly_valid[1] = 1;                     //该区域有效 
    	pBsd_init_para->view_state[1] = VCA_MOD_LEVEL2;          //二级区域
    	pBsd_init_para->bsd_poly[0].vertex_num = 5;
    	pBsd_init_para->bsd_poly[0].point[0].x = (float)1040 / ori_w;
    	pBsd_init_para->bsd_poly[0].point[1].x = (float)1265 / ori_w;
    	pBsd_init_para->bsd_poly[0].point[2].x = (float)1083 / ori_w;
    	pBsd_init_para->bsd_poly[0].point[3].x = (float)413 / ori_w;
    	pBsd_init_para->bsd_poly[0].point[4].x = (float)413 / ori_w;
    
    	pBsd_init_para->bsd_poly[0].point[0].y = (float)218 / ori_h;
    	pBsd_init_para->bsd_poly[0].point[1].y = (float)198 / ori_h;
    	pBsd_init_para->bsd_poly[0].point[2].y = (float)716 / ori_h;
    	pBsd_init_para->bsd_poly[0].point[3].y = (float)716 / ori_h;
    	pBsd_init_para->bsd_poly[0].point[4].y = (float)616 / ori_h;
    
    	//三级区域
    	pBsd_init_para->poly_valid[2] = 1; 					 //该区域有效
    	pBsd_init_para->view_state[2] = VCA_MOD_LEVEL1;			 //三级区域
    	pBsd_init_para->bsd_poly[2].vertex_num = 5;
    	pBsd_init_para->bsd_poly[2].point[0].x = (float)960 / ori_w;	// 925
    	pBsd_init_para->bsd_poly[2].point[1].x = (float)1265 / ori_w; // 
    	pBsd_init_para->bsd_poly[2].point[2].x = (float)1083 / ori_w;
    	pBsd_init_para->bsd_poly[2].point[3].x = (float)183 / ori_w;
    	pBsd_init_para->bsd_poly[2].point[4].x = (float)183 / ori_w;
    
    	pBsd_init_para->bsd_poly[2].point[0].y = (float)198 / ori_h;
    	pBsd_init_para->bsd_poly[2].point[1].y = (float)178 / ori_h;
    	pBsd_init_para->bsd_poly[2].point[2].y = (float)716 / ori_h;
    	pBsd_init_para->bsd_poly[2].point[3].y = (float)716 / ori_h;
    	pBsd_init_para->bsd_poly[2].point[4].y = (float)616 / ori_h;
    
    #endif 
    
    # if 1
    // //右视
    	pBsd_init_para->poly_valid[0] = 1;                     //该区域有效
    	pBsd_init_para->view_state[0] = VCA_MOD_LEVEL3;          //一级区域，最大框
    	pBsd_init_para->bsd_poly[0].vertex_num = 5;
    	pBsd_init_para->bsd_poly[0].point[0].x = (float)431/ori_w;
    	pBsd_init_para->bsd_poly[0].point[0].y = (float)182/ori_h;
    	pBsd_init_para->bsd_poly[0].point[1].x = (float)52/ori_w;
    	pBsd_init_para->bsd_poly[0].point[1].y = (float)163/ori_h;
    	pBsd_init_para->bsd_poly[0].point[2].x = (float)53/ori_w;
    	pBsd_init_para->bsd_poly[0].point[2].y =(float) 720/ori_h;
    	pBsd_init_para->bsd_poly[0].point[3].x =(float) 704/ori_w;
    	pBsd_init_para->bsd_poly[0].point[3].y = (float)712/ori_h;
    	pBsd_init_para->bsd_poly[0].point[4].x = (float)579/ori_w;
    	pBsd_init_para->bsd_poly[0].point[4].y = (float)480/ori_h;
    
    	//二级区域 
    	pBsd_init_para->poly_valid[1] = 1;                     //该区域有效 
    	pBsd_init_para->view_state[1] = VCA_MOD_LEVEL2;          //二级区域
    	pBsd_init_para->bsd_poly[0].vertex_num = 5;
    	pBsd_init_para->bsd_poly[1].point[0].x = (float)298/ori_w;
    	pBsd_init_para->bsd_poly[1].point[0].y = (float)187/ori_h;
    	pBsd_init_para->bsd_poly[1].point[1].x = (float)60/ori_w;
    	pBsd_init_para->bsd_poly[1].point[1].y = (float)172/ori_h;
    	pBsd_init_para->bsd_poly[1].point[2].x = (float)66/ori_w;
    	pBsd_init_para->bsd_poly[1].point[2].y = (float)717/ori_h;
    	pBsd_init_para->bsd_poly[1].point[3].x = (float)481/ori_w;
    	pBsd_init_para->bsd_poly[1].point[3].y = (float)709/ori_h;
    	pBsd_init_para->bsd_poly[1].point[4].x = (float)400/ori_w;
    	pBsd_init_para->bsd_poly[1].point[4].y = (float)495/ori_h;
    
    	//三级区域
    	pBsd_init_para->poly_valid[2] = 1; 					 //该区域有效
    	pBsd_init_para->view_state[2] = VCA_MOD_LEVEL1;			 //三级区域
    	pBsd_init_para->bsd_poly[2].vertex_num = 5;
    	pBsd_init_para->bsd_poly[2].point[0].x = (float)183/ori_w;
    	pBsd_init_para->bsd_poly[2].point[0].y = (float)172/ori_h;
    	pBsd_init_para->bsd_poly[2].point[1].x = (float)75/ori_w;
    	pBsd_init_para->bsd_poly[2].point[1].y = (float)171/ori_h;
    	pBsd_init_para->bsd_poly[2].point[2].x = (float)46/ori_w;
    	pBsd_init_para->bsd_poly[2].point[2].y = (float)705/ori_h;
    	pBsd_init_para->bsd_poly[2].point[3].x = (float)234/ori_w;
    	pBsd_init_para->bsd_poly[2].point[3].y = (float)706/ori_h;
    	pBsd_init_para->bsd_poly[2].point[4].x = (float)218/ori_w;
    	pBsd_init_para->bsd_poly[2].point[4].y = (float)558/ori_h;
    
    #endif
    
    
    
    
    	return OK;
    }
    
    /******************************************************************************* 函数名称: Bsd_draw_jeiguo 描 述: 该函数画图 输入:输入宽高,以及报警框个数 输出: 返回值: 图像 0: Successful ohters: Failed *******************************************************************************/
    Mat Bsd_draw_jeiguo(Mat src, int  width, int height,int lever)
    { 
    	bool OK = 0;/*返回值初始化*/
    	int ori_w = width;
    	int ori_h = height;
    	Mat dst;
    
    	VCA_ADAS_BSDC_MOD_INPUT_PARAM *pBsd_init_para = NULL;
    	pBsd_init_para = &gBsd_init_para;
    	Point p1,p2;
    	/*绘制报警框*/
    	for (int i = 0; i < lever; i++)/*三个等级*/
    	{ 
    		for (int j = 0; j < 5; j++)/*每个报警框由5条线段组成*/
    		{ 
    			p1.x = (int )( pBsd_init_para->bsd_poly[i].point[j].x*ori_w );
    			p1.y = (int )( pBsd_init_para->bsd_poly[i].point[j].y*ori_h );
    			if (j == 4)
    			{ 
    				p2.x = (int )(pBsd_init_para->bsd_poly[i].point[0].x*ori_w);
    				p2.y = (int)( pBsd_init_para->bsd_poly[i].point[0].y*ori_h);
    			}
    			else
    			{ 
    				p2.x = (int)( pBsd_init_para->bsd_poly[i].point[j + 1].x*ori_w);
    				p2.y = (int)( pBsd_init_para->bsd_poly[i].point[j + 1].y*ori_h);
    			}
    			/*绘制直线*/
    			line(src, p1, p2, Scalar(0, 0, 255), 2);
    		}
    	}
    	return src;
    }
    
    /******************************************************************************* 函数名称: Bsd_AlarmInfo 描 述: 判断报警等级 输入:输入报警框, 输出:报警结果 返回值: 图像 0: Successful ohters: Failed *******************************************************************************/
    Mat Bsd_AlarmInfo(Mat src, int  width, int height, int lever)
    { 
    	bool OK = 0;/*返回值初始化*/
    	int ori_w = width;
    	int ori_h = height;
    	Mat dst;
    
    	VCA_ADAS_BSDC_MOD_INPUT_PARAM *pBsd_init_para = NULL;
    	pBsd_init_para = &gBsd_init_para;
    	Point p1, p2;
    	/*绘制报警框*/
    	for (int i = 0; i < lever; i++)/*三个等级*/
    	{ 
    		for (int j = 0; j < 5; j++)/*每个报警框由5条线段组成*/
    		{ 
    			p1.x = (int)(pBsd_init_para->bsd_poly[i].point[j].x*ori_w);
    			p1.y = (int)(pBsd_init_para->bsd_poly[i].point[j].y*ori_h);
    			if (j == 4)
    			{ 
    				p2.x = (int)(pBsd_init_para->bsd_poly[i].point[0].x*ori_w);
    				p2.y = (int)(pBsd_init_para->bsd_poly[i].point[0].y*ori_h);
    			}
    			else
    			{ 
    				p2.x = (int)(pBsd_init_para->bsd_poly[i].point[j + 1].x*ori_w);
    				p2.y = (int)(pBsd_init_para->bsd_poly[i].point[j + 1].y*ori_h);
    			}
    			/*绘制直线*/
    			line(src, p1, p2, Scalar(0, 0, 255), 2);
    		}
    	}
    	return src;
    }
    
    
    
    
    
    
    
    int main()
    { 
    	bool OK = 0;/*返回值初始化*/
    	int ret = TRUE;
    	int BSD_IMAGE_W = 1280;/*宽度*/
    	int BSD_IMAGE_H = 1280;/*高度*/
    	int lever = 1;/*只显示一个报警框*/
    	//string names_file = "coco.names";
    	//string cfg_file = "yolov3-tiny.cfg";
    	//string weights_file = "yolov3-tiny.weights";
    	/*test*/
    	string names_file = "coco.names";
    	string cfg_file = "yolov3-tiny.cfg";
    	string weights_file = "yolov3-tiny.weights";
    
    	//string names_file = "coco.names";
    	//string cfg_file = "yolov4-tiny.cfg";
    	//string weights_file = "yolov4-tiny.weights";
    
    
    
    	Detector detector(cfg_file, weights_file, 0); //初始化检测器
    	vector<string> obj_names;
    	ifstream ifs(names_file.c_str());
    	string line;
    	while (getline(ifs, line)) obj_names.push_back(line);
    	//测试是否成功读入分类对象文件
    	for (size_t i = 0; i < obj_names.size(); i++)
    	{ 
    		cout << obj_names[i] << endl;
    	}
    	/*初始化BSD参数*/
    	ret = BsdInitParam(BSD_IMAGE_W, BSD_IMAGE_H);
    	if (ret != OK)
    	{ 
    		ALG_ERR("配置BSD参数有问题\n");
    	}
    
    	VideoCapture capture;
    	//capture.open("yejianbsd.mp4");
    	capture.open("bsd.mp4");
    	if (!capture.isOpened())
    	{ 
    		printf("文件打开失败");
    	}
    	Mat frame;
    	float fps = 0;
    	while (true)
    	{ 
    		int begin, duration;
    		begin = clock();
    		capture >> frame;
    
    		if (frame.data == NULL)
    		{ 
    			printf("Image is empty");
    			return 0;
    		}
    		
    		duration = clock() - begin;
    		//printf("读图用时约: %d毫秒 \n", duration * 1000 / CLOCKS_PER_SEC);
    		//resize(frame, frame, Size(640,360));
    		vector<bbox_t> result_vec = detector.detect(frame);
    		duration = clock() - begin;
    		printf("识别用时约: %d毫秒 \n", duration * 1000 / CLOCKS_PER_SEC);
    		fps =(float) 1000.0 / duration;
    		draw_boxes(frame, result_vec, obj_names,fps, fps);
    		//duration = clock() - begin;
    		//printf("画框用时约: %d毫秒 \n", duration * 1000 / CLOCKS_PER_SEC);
    		frame = Bsd_draw_jeiguo(frame, BSD_IMAGE_W, BSD_IMAGE_H, lever);
    		namedWindow("bsd", 0);
    		imshow("bsd", frame);
    		waitKey(1);
    		//duration = clock() - begin;
    		//printf("显示用时约: %d毫秒 \n", duration * 1000 / CLOCKS_PER_SEC);
    		printf("\n\n\n");
    	}
    
    	return 0;
    }

##  ##

纯cpu效果图：  
![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L21hb19odWlfZmVp_size_16_color_FFFFFF_t_70 4]

如果只跑cpu，需要把darknet.vcxproj文件中cuda和cudnn路径去掉，不然打不开，  
如下：

<?xml version="1.0" encoding="utf-8"?>
    <Project DefaultTargets="Build" ToolsVersion="14.0" xmlns="http://schemas.microsoft.com/developer/msbuild/2003">
      <ItemGroup Label="ProjectConfigurations">
        <ProjectConfiguration Include="Debug|Win32">
          <Configuration>Debug</Configuration>
          <Platform>Win32</Platform>
        </ProjectConfiguration>
        <ProjectConfiguration Include="Debug|x64">
          <Configuration>Debug</Configuration>
          <Platform>x64</Platform>
        </ProjectConfiguration>
        <ProjectConfiguration Include="Release|Win32">
          <Configuration>Release</Configuration>
          <Platform>Win32</Platform>
        </ProjectConfiguration>
        <ProjectConfiguration Include="Release|x64">
          <Configuration>Release</Configuration>
          <Platform>x64</Platform>
        </ProjectConfiguration>
      </ItemGroup>
      <PropertyGroup Label="Globals">
        <ProjectGuid>{ 4CF5694F-12A5-4012-8D94-9A0915E9FEB5}</ProjectGuid>
        <RootNamespace>darknet</RootNamespace>
        <WindowsTargetPlatformVersion>8.1</WindowsTargetPlatformVersion>
      </PropertyGroup>
      <Import Project="$(VCTargetsPath)\Microsoft.Cpp.Default.props" />
      <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'" Label="Configuration">
        <ConfigurationType>Application</ConfigurationType>
        <UseDebugLibraries>true</UseDebugLibraries>
        <PlatformToolset>v140</PlatformToolset>
        <CharacterSet>MultiByte</CharacterSet>
      </PropertyGroup>
      <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug|x64'" Label="Configuration">
        <ConfigurationType>Application</ConfigurationType>
        <UseDebugLibraries>true</UseDebugLibraries>
        <PlatformToolset>v140</PlatformToolset>
        <CharacterSet>MultiByte</CharacterSet>
      </PropertyGroup>
      <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release|Win32'" Label="Configuration">
        <ConfigurationType>Application</ConfigurationType>
        <UseDebugLibraries>false</UseDebugLibraries>
        <PlatformToolset>v140</PlatformToolset>
        <WholeProgramOptimization>true</WholeProgramOptimization>
        <CharacterSet>MultiByte</CharacterSet>
      </PropertyGroup>
      <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release|x64'" Label="Configuration">
        <ConfigurationType>Application</ConfigurationType>
        <UseDebugLibraries>false</UseDebugLibraries>
        <PlatformToolset>v140</PlatformToolset>
        <WholeProgramOptimization>true</WholeProgramOptimization>
        <CharacterSet>MultiByte</CharacterSet>
      </PropertyGroup>
      <Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
      <ImportGroup Label="ExtensionSettings" />
      <ImportGroup Label="PropertySheets" Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">
        <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" />
      </ImportGroup>
      <ImportGroup Condition="'$(Configuration)|$(Platform)'=='Debug|x64'" Label="PropertySheets">
        <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" />
      </ImportGroup>
      <ImportGroup Label="PropertySheets" Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">
        <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" />
      </ImportGroup>
      <ImportGroup Condition="'$(Configuration)|$(Platform)'=='Release|x64'" Label="PropertySheets">
        <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" />
      </ImportGroup>
      <PropertyGroup Label="UserMacros" />
      <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">
        <OutDir>$(SolutionDir)$(Platform)\</OutDir>
      </PropertyGroup>
      <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release|x64'">
        <OutDir>$(SolutionDir)$(Platform)\</OutDir>
      </PropertyGroup>
      <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">
        <ClCompile>
          <WarningLevel>Level3</WarningLevel>
          <Optimization>Disabled</Optimization>
          <SDLCheck>true</SDLCheck>
        </ClCompile>
        <Link>
          <GenerateDebugInformation>true</GenerateDebugInformation>
        </Link>
      </ItemDefinitionGroup>
      <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">
        <ClCompile>
          <WarningLevel>Level3</WarningLevel>
          <Optimization>Disabled</Optimization>
          <SDLCheck>true</SDLCheck>
          <AdditionalIncludeDirectories>$(OPENCV_DIR)\include;C:\opencv_3.0\opencv\build\include;..\..\include;..\..\3rdparty\stb\include;..\..\3rdparty\pthreads\include;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories>
          <PreprocessorDefinitions>_MBCS;LIB_EXPORTS;_TIMESPEC_DEFINED;_SCL_SECURE_NO_WARNINGS;_CRT_SECURE_NO_WARNINGS;WIN32;NDEBUG;_CONSOLE;_LIB;%(PreprocessorDefinitions)</PreprocessorDefinitions>
          <UndefinePreprocessorDefinitions>CUDNN</UndefinePreprocessorDefinitions>
          <MultiProcessorCompilation>true</MultiProcessorCompilation>
          <ForcedIncludeFiles>stdlib.h;crtdbg.h;%(ForcedIncludeFiles)</ForcedIncludeFiles>
        </ClCompile>
        <Link>
          <GenerateDebugInformation>true</GenerateDebugInformation>
          <AdditionalLibraryDirectories>$(OPENCV_DIR)\x64\vc15\lib;$(OPENCV_DIR)\x64\vc14\lib;C:\opencv_3.0\opencv\build\x64\vc14\lib;..\..\3rdparty\pthreads\lib;%(AdditionalLibraryDirectories)</AdditionalLibraryDirectories>
          <OutputFile>$(OutDir)\$(TargetName)$(TargetExt)</OutputFile>
          <AdditionalDependencies>pthreadVC2.lib;%(AdditionalDependencies)</AdditionalDependencies>
          <AssemblyDebug>true</AssemblyDebug>
        </Link>
        <CudaCompile>
          <CodeGeneration>compute_30,sm_30;compute_75,sm_75</CodeGeneration>
          <TargetMachinePlatform>64</TargetMachinePlatform>
        </CudaCompile>
      </ItemDefinitionGroup>
      <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">
        <ClCompile>
          <WarningLevel>Level3</WarningLevel>
          <Optimization>MaxSpeed</Optimization>
          <FunctionLevelLinking>true</FunctionLevelLinking>
          <IntrinsicFunctions>true</IntrinsicFunctions>
          <SDLCheck>true</SDLCheck>
          <AdditionalIncludeDirectories>C:\opencv_2.4.9\opencv\build\include;..\..\include;..\..\3rdparty\include;%(AdditionalIncludeDirectories);$(CudaToolkitIncludeDir);$(cudnn)\include</AdditionalIncludeDirectories>
          <PreprocessorDefinitions>LIB_EXPORTS;_TIMESPEC_DEFINED;_CRT_SECURE_NO_WARNINGS;GPU;WIN32;NDEBUG;_CONSOLE;_LIB;%(PreprocessorDefinitions)</PreprocessorDefinitions>
          <MultiProcessorCompilation>true</MultiProcessorCompilation>
        </ClCompile>
        <Link>
          <GenerateDebugInformation>true</GenerateDebugInformation>
          <EnableCOMDATFolding>true</EnableCOMDATFolding>
          <OptimizeReferences>true</OptimizeReferences>
          <AdditionalLibraryDirectories>C:\opencv_2.4.9\opencv\build\x86\vc14\lib;C:\opencv_2.4.9\opencv\build\x86\vc12\lib;%(AdditionalLibraryDirectories)</AdditionalLibraryDirectories>
          <AdditionalDependencies>..\..\3rdparty\lib\x86\pthreadVC2.lib;%(AdditionalDependencies)</AdditionalDependencies>
        </Link>
      </ItemDefinitionGroup>
      <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Release|x64'">
        <ClCompile>
          <WarningLevel>Level3</WarningLevel>
          <Optimization>MaxSpeed</Optimization>
          <FunctionLevelLinking>true</FunctionLevelLinking>
          <IntrinsicFunctions>true</IntrinsicFunctions>
          <SDLCheck>true</SDLCheck>
          <AdditionalIncludeDirectories>$(OPENCV_DIR)\include;C:\opencv_3.0\opencv\build\include;..\..\include;..\..\3rdparty\stb\include;..\..\3rdparty\pthreads\include;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories>
          <PreprocessorDefinitions>LIB_EXPORTS;_TIMESPEC_DEFINED;_SCL_SECURE_NO_WARNINGS;_CRT_SECURE_NO_WARNINGS;WIN32;NDEBUG;_CONSOLE;_LIB;%(PreprocessorDefinitions)</PreprocessorDefinitions>
          <CLanguageStandard>c11</CLanguageStandard>
          <CppLanguageStandard>c++1y</CppLanguageStandard>
          <PrecompiledHeaderCompileAs>CompileAsCpp</PrecompiledHeaderCompileAs>
          <CompileAs>Default</CompileAs>
          <UndefinePreprocessorDefinitions>NDEBUG</UndefinePreprocessorDefinitions>
          <MultiProcessorCompilation>true</MultiProcessorCompilation>
          <AdditionalUsingDirectories>
          </AdditionalUsingDirectories>
        </ClCompile>
        <Link>
          <GenerateDebugInformation>true</GenerateDebugInformation>
          <EnableCOMDATFolding>true</EnableCOMDATFolding>
          <OptimizeReferences>true</OptimizeReferences>
          <AdditionalLibraryDirectories>$(OPENCV_DIR)\x64\vc15\lib;$(OPENCV_DIR)\x64\vc14\lib;C:\opencv_3.0\opencv\build\x64\vc14\lib;..\..\3rdparty\pthreads\lib;%(AdditionalLibraryDirectories)</AdditionalLibraryDirectories>
          <AdditionalDependencies>pthreadVC2.lib;%(AdditionalDependencies)</AdditionalDependencies>
          <OutputFile>$(OutDir)\$(TargetName)$(TargetExt)</OutputFile>
        </Link>
        <CudaCompile>
          <TargetMachinePlatform>64</TargetMachinePlatform>
          <CodeGeneration>compute_35,sm_35;compute_86,sm_86</CodeGeneration>
        </CudaCompile>
      </ItemDefinitionGroup>
      <ItemGroup>
        <CudaCompile Include="..\..\src\activation_kernels.cu" />
        <CudaCompile Include="..\..\src\avgpool_layer_kernels.cu" />
        <CudaCompile Include="..\..\src\blas_kernels.cu" />
        <CudaCompile Include="..\..\src\col2im_kernels.cu" />
        <CudaCompile Include="..\..\src\convolutional_kernels.cu" />
        <CudaCompile Include="..\..\src\crop_layer_kernels.cu" />
        <CudaCompile Include="..\..\src\deconvolutional_kernels.cu" />
        <CudaCompile Include="..\..\src\dropout_layer_kernels.cu" />
        <CudaCompile Include="..\..\src\im2col_kernels.cu" />
        <CudaCompile Include="..\..\src\maxpool_layer_kernels.cu" />
        <CudaCompile Include="..\..\src\network_kernels.cu" />
      </ItemGroup>
      <ItemGroup>
        <ClCompile Include="..\..\src\activations.c" />
        <ClCompile Include="..\..\src\activation_layer.c" />
        <ClCompile Include="..\..\src\art.c" />
        <ClCompile Include="..\..\src\avgpool_layer.c" />
        <ClCompile Include="..\..\src\batchnorm_layer.c" />
        <ClCompile Include="..\..\src\blas.c" />
        <ClCompile Include="..\..\src\box.c" />
        <ClCompile Include="..\..\src\captcha.c" />
        <ClCompile Include="..\..\src\cifar.c" />
        <ClCompile Include="..\..\src\classifier.c" />
        <ClCompile Include="..\..\src\coco.c" />
        <ClCompile Include="..\..\src\col2im.c" />
        <ClCompile Include="..\..\src\compare.c" />
        <ClCompile Include="..\..\src\connected_layer.c" />
        <ClCompile Include="..\..\src\convolutional_layer.c" />
        <ClCompile Include="..\..\src\conv_lstm_layer.c" />
        <ClCompile Include="..\..\src\cost_layer.c" />
        <ClCompile Include="..\..\src\cpu_gemm.c" />
        <ClCompile Include="..\..\src\crnn_layer.c" />
        <ClCompile Include="..\..\src\crop_layer.c" />
        <ClCompile Include="..\..\src\darknet.c" />
        <ClCompile Include="..\..\src\dark_cuda.c" />
        <ClCompile Include="..\..\src\data.c" />
        <ClCompile Include="..\..\src\deconvolutional_layer.c" />
        <ClCompile Include="..\..\src\demo.c" />
        <ClCompile Include="..\..\src\detection_layer.c" />
        <ClCompile Include="..\..\src\detector.c" />
        <ClCompile Include="..\..\src\dice.c" />
        <ClCompile Include="..\..\src\dropout_layer.c" />
        <ClCompile Include="..\..\src\gaussian_yolo_layer.c" />
        <ClCompile Include="..\..\src\gemm.c" />
        <ClCompile Include="..\..\src\getopt.c" />
        <ClCompile Include="..\..\src\gettimeofday.c" />
        <ClCompile Include="..\..\src\go.c" />
        <ClCompile Include="..\..\src\gru_layer.c" />
        <ClCompile Include="..\..\src\http_stream.cpp" />
        <ClCompile Include="..\..\src\im2col.c" />
        <ClCompile Include="..\..\src\image.c" />
        <ClCompile Include="..\..\src\image_opencv.cpp" />
        <ClCompile Include="..\..\src\layer.c" />
        <ClCompile Include="..\..\src\list.c" />
        <ClCompile Include="..\..\src\local_layer.c" />
        <ClCompile Include="..\..\src\lstm_layer.c" />
        <ClCompile Include="..\..\src\matrix.c" />
        <ClCompile Include="..\..\src\maxpool_layer.c" />
        <ClCompile Include="..\..\src\network.c" />
        <ClCompile Include="..\..\src\nightmare.c" />
        <ClCompile Include="..\..\src\normalization_layer.c" />
        <ClCompile Include="..\..\src\option_list.c" />
        <ClCompile Include="..\..\src\parser.c" />
        <ClCompile Include="..\..\src\region_layer.c" />
        <ClCompile Include="..\..\src\reorg_layer.c" />
        <ClCompile Include="..\..\src\reorg_old_layer.c" />
        <ClCompile Include="..\..\src\rnn.c" />
        <ClCompile Include="..\..\src\rnn_layer.c" />
        <ClCompile Include="..\..\src\rnn_vid.c" />
        <ClCompile Include="..\..\src\route_layer.c" />
        <ClCompile Include="..\..\src\sam_layer.c" />
        <ClCompile Include="..\..\src\scale_channels_layer.c" />
        <ClCompile Include="..\..\src\shortcut_layer.c" />
        <ClCompile Include="..\..\src\softmax_layer.c" />
        <ClCompile Include="..\..\src\super.c" />
        <ClCompile Include="..\..\src\swag.c" />
        <ClCompile Include="..\..\src\tag.c" />
        <ClCompile Include="..\..\src\tree.c" />
        <ClCompile Include="..\..\src\upsample_layer.c" />
        <ClCompile Include="..\..\src\utils.c" />
        <ClCompile Include="..\..\src\voxel.c" />
        <ClCompile Include="..\..\src\writing.c" />
        <ClCompile Include="..\..\src\yolo.c" />
        <ClCompile Include="..\..\src\yolo_layer.c" />
      </ItemGroup>
      <ItemGroup>
        <ClInclude Include="..\..\include\darknet.h" />
        <ClInclude Include="..\..\src\activations.h" />
        <ClInclude Include="..\..\src\activation_layer.h" />
        <ClInclude Include="..\..\src\avgpool_layer.h" />
        <ClInclude Include="..\..\src\batchnorm_layer.h" />
        <ClInclude Include="..\..\src\blas.h" />
        <ClInclude Include="..\..\src\box.h" />
        <ClInclude Include="..\..\src\classifier.h" />
        <ClInclude Include="..\..\src\col2im.h" />
        <ClInclude Include="..\..\src\connected_layer.h" />
        <ClInclude Include="..\..\src\convolutional_layer.h" />
        <ClInclude Include="..\..\src\conv_lstm_layer.h" />
        <ClInclude Include="..\..\src\cost_layer.h" />
        <ClInclude Include="..\..\src\crnn_layer.h" />
        <ClInclude Include="..\..\src\crop_layer.h" />
        <ClInclude Include="..\..\src\dark_cuda.h" />
        <ClInclude Include="..\..\src\data.h" />
        <ClInclude Include="..\..\src\deconvolutional_layer.h" />
        <ClInclude Include="..\..\src\demo.h" />
        <ClInclude Include="..\..\src\detection_layer.h" />
        <ClInclude Include="..\..\src\dropout_layer.h" />
        <ClInclude Include="..\..\src\gaussian_yolo_layer.h" />
        <ClInclude Include="..\..\src\gemm.h" />
        <ClInclude Include="..\..\src\getopt.h" />
        <ClInclude Include="..\..\src\gettimeofday.h" />
        <ClInclude Include="..\..\src\gru_layer.h" />
        <ClInclude Include="..\..\src\http_stream.h" />
        <ClInclude Include="..\..\src\im2col.h" />
        <ClInclude Include="..\..\src\image.h" />
        <ClInclude Include="..\..\src\image_opencv.h" />
        <ClInclude Include="..\..\src\layer.h" />
        <ClInclude Include="..\..\src\list.h" />
        <ClInclude Include="..\..\src\local_layer.h" />
        <ClInclude Include="..\..\src\lstm_layer.h" />
        <ClInclude Include="..\..\src\matrix.h" />
        <ClInclude Include="..\..\src\maxpool_layer.h" />
        <ClInclude Include="..\..\src\network.h" />
        <ClInclude Include="..\..\src\normalization_layer.h" />
        <ClInclude Include="..\..\src\option_list.h" />
        <ClInclude Include="..\..\src\parser.h" />
        <ClInclude Include="..\..\src\region_layer.h" />
        <ClInclude Include="..\..\src\reorg_layer.h" />
        <ClInclude Include="..\..\src\reorg_old_layer.h" />
        <ClInclude Include="..\..\src\rnn_layer.h" />
        <ClInclude Include="..\..\src\route_layer.h" />
        <ClInclude Include="..\..\src\sam_layer.h" />
        <ClInclude Include="..\..\src\scale_channels_layer.h" />
        <ClInclude Include="..\..\src\shortcut_layer.h" />
        <ClInclude Include="..\..\src\softmax_layer.h" />
        <ClInclude Include="..\..\src\stb_image.h" />
        <ClInclude Include="..\..\src\stb_image_write.h" />
        <ClInclude Include="..\..\src\tree.h" />
        <ClInclude Include="..\..\src\unistd.h" />
        <ClInclude Include="..\..\src\upsample_layer.h" />
        <ClInclude Include="..\..\src\utils.h" />
        <ClInclude Include="..\..\src\yolo_layer.h" />
      </ItemGroup>
      <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
      <ImportGroup Label="ExtensionTargets" />
    </Project>

## 封装及调用——方便工程化。 ##

[Windows系统下YOLO动态链接库的封装和调用(Windows10+VS2015+OpenCV3.4.0+CUDA9.0+cuDNN7.0)][https_blog.csdn.net_stjuliet_article_details_87884976]

https://blog.csdn.net/weixin\_42783784/article/details/105741982

https://blog.csdn.net/Schling/article/details/114597758

其他参考

**官方网址**：[https://github.com/hgpvision/darknet][https_github.com_hgpvision_darknet]

2、[YOLO3 darknet训练自己的数据][YOLO3 darknet]

[详细步骤！使用yolov3-tiny训练，测试、验证VOC数据集][yolov3-tiny_VOC 1]

[https_github.com_pjreddie_darknet]: https://github.com/pjreddie/darknet
[https_pjreddie.com_media_files_yolov3_tiny.weights]: https://pjreddie.com/media/files/yolov3_tiny.weights
[yolov3-tiny]: https://download.csdn.net/download/1984/10822410?utm_source=bbsseo
[yolov3-tiny_VOC]: https://blog.csdn.net/HuanglnQuan/article/details/103602118?utm_medium=distribute.pc_relevant.none-task-blog-BlogCommendFromMachineLearnPai2-1.control&dist_request_id=1328593.24855.16148395715075967&depth_1-utm_source=distribute.pc_relevant.none-task-blog-BlogCommendFromMachineLearnPai2-1.control
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L21hb19odWlfZmVp_size_16_color_FFFFFF_t_70]: /images/20221024/8760f7d101cd47c1a9b504767f05e4ab.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L21hb19odWlfZmVp_size_16_color_FFFFFF_t_70 1]: /images/20221024/a59b317145e24146bdc3a99aad70e4b1.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L21hb19odWlfZmVp_size_16_color_FFFFFF_t_70 2]: /images/20221024/c3875b3787e44325ae37425e545f3f7c.png
[https_blog.csdn.net_qq_28663849_article_details_107362445]: https://blog.csdn.net/qq_28663849/article/details/107362445
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L21hb19odWlfZmVp_size_16_color_FFFFFF_t_70 3]: /images/20221024/34ab2b6a7a90498eb02ca8fb91df2140.png
[https_blog.csdn.net_stjuliet_article_details_87884976]: https://blog.csdn.net/stjuliet/article/details/87884976
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L21hb19odWlfZmVp_size_16_color_FFFFFF_t_70 4]: /images/20221024/06eb797427f542c08a8c21504c5001f6.png
[https_github.com_hgpvision_darknet]: https://github.com/hgpvision/darknet
[YOLO3 darknet]: https://zhuanlan.zhihu.com/p/45852709
[yolov3-tiny_VOC 1]: https://blog.csdn.net/HuanglnQuan/article/details/103602118?utm_medium=distribute.pc_relevant.none-task-blog-BlogCommendFromMachineLearnPai2-1.control&dist_request_id=1328592.13871.16147751965791459&depth_1-utm_source=distribute.pc_relevant.none-task-blog-BlogCommendFromMachineLearnPai2-1.control