Application of object detection based on OpenCV and yolov3

One 、 Preface

YOLOv3 It is a very popular target detection algorithm , Its detection speed is very fast , At the same time, the accuracy is also relatively high , Suitable for industrial applications . On the Internet , There are a lot of blog posts to introduce YOLOv3 The basic principle of , therefore , In this article, there is no longer too much description of the principle , Next we'll learn how to OpenCV Use the current more advanced target detection technology YOLOv3 To detect the target in the image .

Two 、 application

In our usual target detection , There are two common ways to detect objects: picture and video stream , However, this platform only supports Tencent video 、 Youku Video 、 BiliBili video upload , Therefore, the following only gives the use YOLOv3 The test of image by target detection algorithm .
First find a picture of the same kind of target to test , give the result as follows ：
Before testing ：

After testing

Then test a picture with different categories ：
Before testing ：

After testing ：

3、 ... and 、 summary

It can be seen from the figure after the above test , Use YOLOv3 The target detection algorithm has a high detection rate for the objects in the picture and the accuracy after detection , The real-time performance shown in the upper left corner of the image is also faster than other target detection algorithms .