OpenCV+yolov3实现目标检测(C++,Python)
OpenCV+yolov3实现目标检测(C++,Python)
目标检测算法主要分为两类:一类是基于Region Proposal(候选区域)的算法,如R-CNN系算法(R-CNN,Fast R-CNN, Faster R-CNN),它们是two-stage(两步法)的,需要先使用Selective search或者CNN网络(RPN)产生Region Proposal,然后再在Region Proposal上做分类与回归。而另一类是Yolo,SSD这类one-stage算法(一步法),其仅仅使用一个CNN网络直接预测不同目标的类别与位置。第一类方法是准确度高一些,但是速度慢,而第二类算法是速度快,但是准确性要低一些。
YOLO是一种比SSD还要快的目标检测网络模型,作者在其论文中说FPS是Fast R-CNN的100倍,这里首先简单的介绍一下YOLO网络基本结构,然后通过OpenCV C++调用Darknet的,实现目标检测。OpenCV在3.3.1的版本中开始正式支持Darknet网络框架并且支持YOLO1与YOLO2以及YOLO Tiny网络模型的导入与使用。后面测试,OpenCV3.4.2也支持YOLO3 。另外,OpenCV dnn模块目前支持Caffe、TensorFlow、Torch、PyTorch等深度学习框架,关于《OpenCV调用TensorFlow预训练模型》可参考鄙人的另一份博客:https://blog.csdn.net/guyuealian/article/details/80570120
关于《OpenCV+yolov2-tiny实现目标检测(C++)》请参考我的另一篇博客:https://blog.csdn.net/guyuealian/article/details/82950283
本博客源码都放在Github上:https://github.com/PanJinquan/opencv-learning-tutorials/tree/master/dnn_tutorial,麻烦给个“Star”哈
参考资料:
《Deep Learning based Object Detection using YOLOv3 with OpenCV ( Python / C++ )》:
官网博客:https://www.learnopencv.com/deep-learning-based-object-detection-using-yolov3-with-opencv-python-c/
《YOLOv3 + OpenCV 实现目标检测(Python / C ++)》:https://blog.csdn.net/haoqimao_hard/article/details/82081285
Github参考源码:https://github.com/spmallick/learnopencv/tree/master/ObjectDetection-YOLO
darknt yolo官网:https://pjreddie.com/darknet/yolo/
目录
OpenCV+yolov3实现目标检测(C++,Python)
1、YOLO网络
(1)YOLO网络结构
2、OpenCV使用YOLO v3实现目标检测
2.1 C++代码
2.2 Python代码
3、YOLO的缺点
4、参考资料:
1、YOLOv3网络
相比YOLOv2和YOLOv1,YOLOv3最大的变化包括两点:使用残差模型和采用FPN架构。YOLOv3的特征提取器是一个残差模型,因为包含53个卷积层,所以称为Darknet-53,从网络结构上看,相比Darknet-19网络使用了残差单元,所以可以构建得更深。另外一个点是采用FPN架构(Feature Pyramid Networks for Object Detection)来实现多尺度检测
YOLOv3是到目前为止,速度和精度最均衡的目标检测网络。通过多种先进方法的融合,将YOLO系列的短板(速度很快,不擅长检测小物体等)全部补齐。
1.1 YOLOv3网络结构
参考资料:
《深入理解目标检测与YOLO(从v1到v3)》:https://blog.csdn.net/qq_39521554/article/details/80694512
https://blog.csdn.net/leviopku/article/details/82660381
2、OpenCV使用YOLO v3实现目标检测
yolov3模型下载地址:链接: https://pan.baidu.com/s/1TugNSWRZaJz1R6IejRtNiA 提取码: 46mh
2.1 C++代码
// This code is written at BigVision LLC. It is based on the OpenCV project.
//It is subject to the license terms in the LICENSE file found in this distribution and at http://opencv.org/license.html// Usage example: ./object_detection_yolo.out --video=run.mp4
// ./object_detection_yolo.out --image=bird.jpg
#include <fstream>
#include <sstream>
#include <iostream>#include <opencv2/dnn.hpp>
#include <opencv2/imgproc.hpp>
#include <opencv2/highgui.hpp>
;
using namespace cv;
using namespace dnn;
using namespace std;string pro_dir = "E:/opencv-learning-tutorials/"; //项目根目录// Initialize the parameters
float confThreshold = 0.5; // Confidence threshold
float nmsThreshold = 0.4; // Non-maximum suppression threshold
int inpWidth = 416; // Width of network's input image
int inpHeight = 416; // Height of network's input image
vector<string> classes;// Remove the bounding boxes with low confidence using non-maxima suppression
void postprocess(Mat& frame, const vector<Mat>& out);// Draw the predicted bounding box
void drawPred(int classId, float conf, int left, int top, int right, int bottom, Mat& frame);// Get the names of the output layers
vector<String> getOutputsNames(const Net& net);void detect_image(string image_path, string modelWeights, string modelConfiguration, string classesFile);void detect_video(string video_path, string modelWeights, string modelConfiguration, string classesFile);int main(int argc, char** argv)
{// Give the configuration and weight files for the modelString modelConfiguration = pro_dir + "data/models/yolov3/yolov3.cfg";String modelWeights = pro_dir + "data/models/yolov3/yolov3.weights";string image_path = pro_dir + "data/images/bird.jpg";string classesFile = pro_dir + "data/models/yolov3/coco.names";// "coco.names";//detect_image(image_path, modelWeights, modelConfiguration, classesFile);string video_path = pro_dir + "data/images/run.mp4";detect_video(video_path, modelWeights, modelConfiguration, classesFile);cv::waitKey(0);return 0;
}void detect_image(string image_path, string modelWeights, string modelConfiguration, string classesFile) {// Load names of classesifstream ifs(classesFile.c_str());string line;while (getline(ifs, line)) classes.push_back(line);// Load the networkNet net = readNetFromDarknet(modelConfiguration, modelWeights);net.setPreferableBackend(DNN_BACKEND_OPENCV);net.setPreferableTarget(DNN_TARGET_OPENCL);// Open a video file or an image file or a camera stream.string str, outputFile;cv::Mat frame = cv::imread(image_path);// Create a windowstatic const string kWinName = "Deep learning object detection in OpenCV";namedWindow(kWinName, WINDOW_NORMAL);// Stop the program if reached end of video// Create a 4D blob from a frame.Mat blob;blobFromImage(frame, blob, 1 / 255.0, cvSize(inpWidth, inpHeight), Scalar(0, 0, 0), true, false);//Sets the input to the networknet.setInput(blob);// Runs the forward pass to get output of the output layersvector<Mat> outs;net.forward(outs, getOutputsNames(net));// Remove the bounding boxes with low confidencepostprocess(frame, outs);// Put efficiency information. The function getPerfProfile returns the overall time for inference(t) and the timings for each of the layers(in layersTimes)vector<double> layersTimes;double freq = getTickFrequency() / 1000;double t = net.getPerfProfile(layersTimes) / freq;string label = format("Inference time for a frame : %.2f ms", t);putText(frame, label, Point(0, 15), FONT_HERSHEY_SIMPLEX, 0.5, Scalar(0, 0, 255));// Write the frame with the detection boxesimshow(kWinName, frame);cv::waitKey(30);
}void detect_video(string video_path, string modelWeights, string modelConfiguration, string classesFile) {string outputFile = "./yolo_out_cpp.avi";;// Load names of classesifstream ifs(classesFile.c_str());string line;while (getline(ifs, line)) classes.push_back(line);// Load the networkNet net = readNetFromDarknet(modelConfiguration, modelWeights);net.setPreferableBackend(DNN_BACKEND_OPENCV);net.setPreferableTarget(DNN_TARGET_CPU);// Open a video file or an image file or a camera stream.VideoCapture cap;//VideoWriter video;Mat frame, blob;try {// Open the video fileifstream ifile(video_path);if (!ifile) throw("error");cap.open(video_path);}catch (...) {cout << "Could not open the input image/video stream" << endl;return ;}// Get the video writer initialized to save the output video//video.open(outputFile, // VideoWriter::fourcc('M', 'J', 'P', 'G'), // 28, // Size(cap.get(CAP_PROP_FRAME_WIDTH), cap.get(CAP_PROP_FRAME_HEIGHT)));// Create a windowstatic const string kWinName = "Deep learning object detection in OpenCV";namedWindow(kWinName, WINDOW_NORMAL);// Process frames.while (waitKey(1) < 0){// get frame from the videocap >> frame;// Stop the program if reached end of videoif (frame.empty()) {cout << "Done processing !!!" << endl;cout << "Output file is stored as " << outputFile << endl;waitKey(3000);break;}// Create a 4D blob from a frame.blobFromImage(frame, blob, 1 / 255.0, cvSize(inpWidth, inpHeight), Scalar(0, 0, 0), true, false);//Sets the input to the networknet.setInput(blob);// Runs the forward pass to get output of the output layersvector<Mat> outs;net.forward(outs, getOutputsNames(net));// Remove the bounding boxes with low confidencepostprocess(frame, outs);// Put efficiency information. The function getPerfProfile returns the overall time for inference(t) and the timings for each of the layers(in layersTimes)vector<double> layersTimes;double freq = getTickFrequency() / 1000;double t = net.getPerfProfile(layersTimes) / freq;string label = format("Inference time for a frame : %.2f ms", t);putText(frame, label, Point(0, 15), FONT_HERSHEY_SIMPLEX, 0.5, Scalar(0, 0, 255));// Write the frame with the detection boxesMat detectedFrame;frame.convertTo(detectedFrame, CV_8U);//video.write(detectedFrame);imshow(kWinName, frame);}cap.release();//video.release();}// Remove the bounding boxes with low confidence using non-maxima suppression
void postprocess(Mat& frame, const vector<Mat>& outs)
{vector<int> classIds;vector<float> confidences;vector<Rect> boxes;for (size_t i = 0; i < outs.size(); ++i){// Scan through all the bounding boxes output from the network and keep only the// ones with high confidence scores. Assign the box's class label as the class// with the highest score for the box.float* data = (float*)outs[i].data;for (int j = 0; j < outs[i].rows; ++j, data += outs[i].cols){Mat scores = outs[i].row(j).colRange(5, outs[i].cols);Point classIdPoint;double confidence;// Get the value and location of the maximum scoreminMaxLoc(scores, 0, &confidence, 0, &classIdPoint);if (confidence > confThreshold){int centerX = (int)(data[0] * frame.cols);int centerY = (int)(data[1] * frame.rows);int width = (int)(data[2] * frame.cols);int height = (int)(data[3] * frame.rows);int left = centerX - width / 2;int top = centerY - height / 2;classIds.push_back(classIdPoint.x);confidences.push_back((float)confidence);boxes.push_back(Rect(left, top, width, height));}}}// Perform non maximum suppression to eliminate redundant overlapping boxes with// lower confidencesvector<int> indices;NMSBoxes(boxes, confidences, confThreshold, nmsThreshold, indices);for (size_t i = 0; i < indices.size(); ++i){int idx = indices[i];Rect box = boxes[idx];drawPred(classIds[idx], confidences[idx], box.x, box.y,box.x + box.width, box.y + box.height, frame);}
}// Draw the predicted bounding box
void drawPred(int classId, float conf, int left, int top, int right, int bottom, Mat& frame)
{//Draw a rectangle displaying the bounding boxrectangle(frame, Point(left, top), Point(right, bottom), Scalar(255, 178, 50), 3);//Get the label for the class name and its confidencestring label = format("%.2f", conf);if (!classes.empty()){CV_Assert(classId < (int)classes.size());label = classes[classId] + ":" + label;}//Display the label at the top of the bounding boxint baseLine;Size labelSize = getTextSize(label, FONT_HERSHEY_SIMPLEX, 0.5, 1, &baseLine);top = max(top, labelSize.height);rectangle(frame, Point(left, top - round(1.5*labelSize.height)), Point(left + round(1.5*labelSize.width), top + baseLine), Scalar(255, 255, 255), FILLED);putText(frame, label, Point(left, top), FONT_HERSHEY_SIMPLEX, 0.75, Scalar(0, 0, 0), 1);
}// Get the names of the output layers
vector<String> getOutputsNames(const Net& net)
{static vector<String> names;if (names.empty()){//Get the indices of the output layers, i.e. the layers with unconnected outputsvector<int> outLayers = net.getUnconnectedOutLayers();//get the names of all the layers in the networkvector<String> layersNames = net.getLayerNames();// Get the names of the output layers in namesnames.resize(outLayers.size());for (size_t i = 0; i < outLayers.size(); ++i)names[i] = layersNames[outLayers[i] - 1];}return names;
}
2.2 Python代码
使用cv_dnn_forward获得预测输出outs是三个二维的数组,每个二维数组是一个feature_map的输出结果,feature_map中每一行是一个预测值:
outs:[507*85 =13*13*3*(5+80),
2028*85=26*26*3*(5+80),
8112*85=52*52*3*(5+80)]每一个行:85=[x,y,w,h,confs,class_probs_0,class_probs_1,..,class_probs_78,class_probs_79]
# -*-coding: utf-8 -*-
"""@Project: tensorflow-yolov3@File : opencv_dnn_yolov3.py@Author : panjq@E-mail : pan_jinquan@163.com@Date : 2019-01-28 14:36:00
"""import cv2 as cv
import numpy as npdef read_class(file):with open(file, 'rt') as f:classes = f.read().rstrip('\n').split('\n')return classesclass cv_yolov3(object):def __init__(self,class_path,net_width,net_height,confThreshold,nmsThreshold):'''Initialize the parameters:param class_path::param net_width: default 416, Width of network's input image:param net_height: default 416,Height of network's input image:param confThreshold: default 0.5, Confidence threshold:param nmsThreshold: default 0.5,Non-maximum suppression threshold'''self.classes = read_class(class_path)self.net_width=net_widthself.net_height=net_heightself.confThreshold=confThresholdself.nmsThreshold=nmsThresholddef cv_dnn_init(self,modelConfiguration,modelWeights):'''Give the configuration and weight files for the model and load the network using them.eg:modelConfiguration = "checkpoint-bk/yolov3.cfg";modelWeights = "checkpoint-bk/yolov3.weights";:param modelConfiguration::param modelWeights::return:'''self.net = cv.dnn.readNetFromDarknet(modelConfiguration, modelWeights)self.net.setPreferableBackend(cv.dnn.DNN_BACKEND_OPENCV)self.net.setPreferableTarget(cv.dnn.DNN_TARGET_CPU)def getOutputsNames(self,net):'''Get the names of the output layers:param net::return:'''# Get the names of all the layers in the networklayersNames = net.getLayerNames()# Get the names of the output layers, i.e. the layers with unconnected outputsreturn [layersNames[i[0] - 1] for i in net.getUnconnectedOutLayers()]def drawPred(self,frame,classes,classId, conf, left, top, right, bottom):'''Draw the predicted bounding box:param frame::param classes::param classId::param conf::param left::param top::param right::param bottom::return:'''# Draw a bounding box.cv.rectangle(frame, (left, top), (right, bottom), (255, 178, 50), 3)label = '%.2f' % conf# Get the label for the class name and its confidenceif classes:assert (classId < len(classes))label = '%s:%s' % (classes[classId], label)# Display the label at the top of the bounding boxlabelSize, baseLine = cv.getTextSize(label, cv.FONT_HERSHEY_SIMPLEX, 0.5, 1)top = max(top, labelSize[1])cv.rectangle(frame, (left, top - round(1.5 * labelSize[1])), (left + round(1.5 * labelSize[0]), top + baseLine),(255, 255, 255), cv.FILLED)cv.putText(frame, label, (left, top), cv.FONT_HERSHEY_SIMPLEX, 0.75, (0, 0, 0), 1)def postprocess(self,frame, classes,outs):'''Remove the bounding boxes with low confidence using non-maxima suppression:param frame::param classes::return: outs:[507*85 =(13*13*3)*(5+80),2028*85=(26*26*3)*(5+80),8112*85=(52*52*3)*(5+80)]outs中每一行是一个预测值:[x,y,w,h,confs,class_probs_0,class_probs_1,..,class_probs_78,class_probs_79]:return:'''frameHeight = frame.shape[0]frameWidth = frame.shape[1]# Scan through all the bounding boxes output from the network and keep only the# ones with high confidence scores. Assign the box's class label as the class with the highest score.classIds = []confidences = []boxes = []for out in outs:for detection in out:scores = detection[5:]classId = np.argmax(scores)confidence = scores[classId]if confidence > self.confThreshold:center_x = int(detection[0] * frameWidth)center_y = int(detection[1] * frameHeight)width = int(detection[2] * frameWidth)height = int(detection[3] * frameHeight)left = int(center_x - width / 2)top = int(center_y - height / 2)classIds.append(classId)confidences.append(float(confidence))boxes.append([left, top, width, height])# Perform non maximum suppression to eliminate redundant overlapping boxes with# lower confidences.indices = cv.dnn.NMSBoxes(boxes, confidences, self.confThreshold, self.nmsThreshold)for i in indices:i = i[0]box = boxes[i]left = box[0]top = box[1]width = box[2]height = box[3]self.drawPred(frame,classes,classIds[i], confidences[i], left, top, left + width, top + height)def cv_dnn_forward(self,frame):''':param frame::return: outs:[507*85 =13*13*3*(5+80),2028*85=26*26*3*(5+80),8112*85=52*52*3*(5+80)]'''# Create a 4D blob from a frame.blob = cv.dnn.blobFromImage(frame, 1 / 255, (self.net_width, self.net_height), [0, 0, 0], 1, crop=False)# Sets the input to the networkself.net.setInput(blob)# Runs the forward pass to get output of the output layersouts = self.net.forward(self.getOutputsNames(self.net))# Put efficiency information. The function getPerfProfile returns the overall time for inference(t) and the timings for each of the layers(in layersTimes)runtime, _ = self.net.getPerfProfile()return outs,runtimedef yolov3_predict(self,image_path):''':param image_path::return:'''# Process inputswinName = 'Deep learning object detection in OpenCV'cv.namedWindow(winName, cv.WINDOW_NORMAL)frame=cv.imread(image_path)outs,runtime=self.cv_dnn_forward(frame)# Remove the bounding boxes with low confidenceself.postprocess(frame, self.classes, outs)label = 'Inference time: %.2f ms' % (runtime * 1000.0 / cv.getTickFrequency())cv.putText(frame, label, (0, 15), cv.FONT_HERSHEY_SIMPLEX, 0.5, (0, 0, 255))cv.imshow(winName, frame)cv.waitKey(0)if __name__=="__main__":confThreshold = 0.5 # Confidence thresholdnmsThreshold = 0.5 # Non-maximum suppression thresholdnet_input_width = 416 # Width of network's input imagenet_input_height = 416 # Height of network's input imageimage_path = "./data/demo_data/dog.jpg"# anchors_path = './data/coco_anchors.txt'classesFile = './data/coco.names'modelConfiguration = "model/yolov3.cfg";modelWeights = "model/yolov3.weights";cv_model=cv_yolov3(classesFile,net_input_width,net_input_height,confThreshold,nmsThreshold)cv_model.cv_dnn_init(modelConfiguration,modelWeights)cv_model.yolov3_predict(image_path)
3、YOLO的缺点
- YOLO对相互靠的很近的物体,还有很小的群体 检测效果不好,这是因为一个网格中只预测了两个框,并且只属于一类。
- 对测试图像中,同一类物体出现的新的不常见的长宽比和其他情况是。泛化能力偏弱。
- 由于损失函数的问题,定位误差是影响检测效果的主要原因。尤其是大小物体的处理上,还有待加强。
4、参考资料:
[1].《论文阅读笔记:You Only Look Once: Unified, Real-Time Object Detection》https://blog.csdn.net/tangwei2014/article/details/50915317
[2]. https://blog.csdn.net/xiaohu2022/article/details/79211732
[3]. https://blog.csdn.net/u014380165/article/details/72616238
OpenCV+yolov3实现目标检测(C++,Python)相关推荐
- opencv+yolov3实现目标检测
效果图 代码/font> 语言:python import cv2 as cv import argparse import numpy as np# Initialize the parame ...
- OpenCV+yolov2-tiny实现目标检测(C++)
OpenCV+yolov2-tiny实现目标检测(C++) 目标检测算法主要分为两类:一类是基于Region Proposal(候选区域)的算法,如R-CNN系算法(R-CNN,Fast R-CNN, ...
- python简单目标检测代码_Python Opencv实现单目标检测的示例代码
一 简介 目标检测即为在图像中找到自己感兴趣的部分,将其分割出来进行下一步操作,可避免背景的干扰.以下介绍几种基于opencv的单目标检测算法,算法总体思想先尽量将目标区域的像素值全置为1,背景区域全 ...
- 目标检测实战必会!4种基于YOLO目标检测(Python和C++两种版本实现)
目标检测实战必会!4种基于YOLO目标检测(Python和C++两种版本实现) AI算法修炼营 1周前 以下文章来源于极市平台 ,作者CV开发者都爱看的 极市平台 专注计算机视觉前沿资讯和技术干货,官 ...
- 【教程】opencv-python+yolov3实现目标检测
目录 [教程]opencv-python+yolov3实现目标检测 目标检测概况 目标检测是? 目标检测算法? yolov3模型简介 性能介绍 架构介绍 opencv-python实现 why ope ...
- 智慧交通day02-车流量检测实现12:基于yoloV3的目标检测
在本章节代码编写中,发现之前的代码所处的环境是python3,因此导致了cv2.dnn.readNetFromDarknet()在代码运行中导致了i[0]的获值失败,故总结如下: cv2.dnn.re ...
- 【Yolo3】入门目标检测实验--Python+Opencv2+dnn
文章目录 一.前言 二.入门实验coco数据集 1.下载 2.食用 3.源码 一.前言 yolo3也是目标检测的新兴算法之一.它的发展是基于HOG->CNN ->RCNN->YOLO ...
- YOLOv3物体/目标检测之实战篇(Windows系统、Python3、TensorFlow2版本)
前言 基于YOLO进行物体检测.对象识别,在搭建好开发环境后,先和大家进行实践应用中,体验YOLOv3物体/目标检测效果和魅力:同时逐步了解YOLOv3的不足和优化思路. 开发环境参数 系统:Wind ...
- 用C++ 和OpenCV 实现视频目标检测(YOLOv4模型)
点击上方"视学算法",选择加"星标"或"置顶" 重磅干货,第一时间送达 据说,现在很多小区都上线了AI抓拍高空抛物的黑科技,可以自动分析抛物 ...
最新文章
- 使用KNN模型进行多输出分类实战(Multioutput Classification)
- CF396C On Changing Tree
- 简明python教程 豆瓣-Python 有哪些入门学习方法和值得推荐的经典教材?
- Maven私服搭建(Nexus Repository Manager 3)
- 插件~NuGet与packages管理项目的包包
- tomcat 页面图片 管理 服务器 配置
- 消息订阅与发布(pubsub)
- Docker(二十三)-Docker使用pipework配置本地网络
- 基础算法---二分查找
- js,jquery小知识点
- java经典编程练习题_java笔试经典练习题及答案
- U盘_PE启动+U存储+kali三合一启动盘制作教程
- netware显示没有首选服务器,NetWare下服务器配置几例
- 程序虽小但功能强大!5个实用的微信小程序,每款都是生活好帮手
- 南方电网数据安全建设分析——云集至
- 基于SpringBoot+VUE(PC端+小程序端)的智能在线考试系统毕业设计
- 嵌入式端音频开发(实战篇)之 4.2 把很多音频合成一个bin文件并统计每个音频文件的大小的方法(内含转换脚本)
- 服务器系统关闭显卡加速,Windows Server2008 R2性能优化方法
- 【艾特淘】店铺没流量?教你如何获取免费流量来源
- java图书销售系统_基于springboot的小型图书销售系统 源码下载