use-yolov3-with-opencv-in-python.html

软硬件环境

ubuntu 18.04 64bit
NVIDIA GTX 1070Ti 8G
anaconda with python 3.6
opencv 3.4.3
cuda 9.0
YOLO v3

前言

下图是近年来物体检测领域算法的演化，YOLO是目前公认的比较准确的物体检测算法，已经发展到了第三个版本。关于darknet(实现YOLO检测的开源项目)的基本情况，参考之前的博文 https://blog.xugaoxiang.com/ai/darknet.html，里面有比较详细的阐述。

准备工作

下载YOLO检测需要用到的配置文件、weights模型文件及物体类型class文件

wget https://pjreddie.com/media/files/yolov3.weights
wget https://github.com/pjreddie/darknet/blob/master/cfg/yolov3.cfg?raw=true -O ./yolov3.cfg
wget https://github.com/pjreddie/darknet/blob/master/data/coco.names?raw=true -O ./coco.names

YOLO的基本原理

一般来讲，物体检测由两部分组成，物体定位(object locator)和物体识别(object recognizer)。下面以图片为例来讲YOLO的实现原理

将图片分割成13x13大小的网格单元, 以一张416x416像素大小的图片为例，会有1024个网格单元，模型会在每个cell上预测bounding box
每个cell有可能会被预测出多个bounding box,大部分bounding box最后都会被清除，因为它们的相似度太低。这里有个算法叫Non-Maximum Suppression，翻译过来叫非极大值抑制，可参考论文Efficient-Non-Maximum-Suppression,　NMS算法是用来提取相似度最高的

python代码

opencv 3.4.2及以上的版本已经支持darknet，同时也支持使用其他常见深度学习框架的模型，如torch、tensorflow、caffe等。

# -*- coding: utf-8 -*-
# @time    : 18-10-26 下午4:47
# @author  : xugaoxiang
# @email   : djstava@gmail.com
# @website : https://xugaoxiang.com
# @file    : opencv_yolov3.py
# @software: PyCharm# Usage example: python3 opencv_yolov3.py --image=test.pngimport sys
import cv2
import argparse
import numpy as np
import os.path# 参数初始化
# 相似度阈值
confThreshold = 0.5  # Confidence threshold# NMS算法阈值
nmsThreshold = 0.4# 输入图片的宽和高
inpWidth = 416
inpHeight = 416parser = argparse.ArgumentParser(description = 'Object detection using YOLOv3 in opencv')
parser.add_argument('--image', help = 'Path to image file.')
args = parser.parse_args()# 导入物体类别class文件，默认支持80种
classesFile = "coco.names"
classes = None
with open(classesFile, 'rt') as f :classes = f.read().rstrip('\n').split('\n')# yolo v3的配置及weights文件
modelConfiguration = "yolov3.cfg"
modelWeights = "yolov3.weights"# opencv读取外部模型
net = cv2.dnn.readNetFromDarknet(modelConfiguration, modelWeights)
net.setPreferableBackend(cv2.dnn.DNN_BACKEND_OPENCV)
# 这里使用CPU，如果想使用GPU的话，参数是DNN_TARGET_OPENCL,　但是当前版本只支持interl GPU，如果是其它GPU的话，会自动切换到CPU模式
net.setPreferableTarget(cv2.dnn.DNN_TARGET_CPU)# Get the names of the output layers
def getOutputsNames(net) :# 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()]# 画bounding box
def drawPred(classId, conf, left, top, right, bottom) :# Draw a bounding box.cv2.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 = cv2.getTextSize(label, cv2.FONT_HERSHEY_SIMPLEX, 0.5, 1)top = max(top, labelSize[1])cv2.rectangle(frame, (left, top - round(1.5 * labelSize[1])), (left + round(1.5 * labelSize[0]), top + baseLine),(255, 255, 255), cv2.FILLED)cv2.putText(frame, label, (left, top), cv2.FONT_HERSHEY_SIMPLEX, 0.75, (0, 0, 0), 1)# 使用NMS算法，丢弃低相似度的bounding box
def postprocess(frame, outs) :frameHeight = frame.shape[0]frameWidth = frame.shape[1]classIds = []confidences = []boxes = []# 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 > 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 = cv2.dnn.NMSBoxes(boxes, confidences, confThreshold, nmsThreshold)for i in indices :i = i[0]box = boxes[i]left = box[0]top = box[1]width = box[2]height = box[3]drawPred(classIds[i], confidences[i], left, top, left + width, top + height)# Process inputs
winName = 'Deep learning object detection in OpenCV'
cv2.namedWindow(winName, cv2.WINDOW_NORMAL)if (args.image) :if not os.path.isfile(args.image) :print('Input image file {} does not exist.'.format(args.image))sys.exit(1)frame = cv2.imread(args.image, cv2.IMREAD_ANYCOLOR)outputFile = args.image[:-4] + '_yolov3_out.png'# Create a 4D blob from a frame.blob = cv2.dnn.blobFromImage(frame, 1 / 255, (inpWidth, inpHeight), [0, 0, 0], 1, crop = False)# Sets the input to the networknet.setInput(blob)# Runs the forward pass to get output of the output layersouts = net.forward(getOutputsNames(net))# Remove the bounding boxes with low confidencepostprocess(frame, outs)cv2.imshow(winName, frame)cv2.imwrite(outputFile, frame)cv2.destroyAllWindows()

测试程序输出

参考资料

https://github.com/pjreddie/darknet/wiki/YOLO:-Real-Time-Object-Detection
https://docs.opencv.org/3.4.3/da/d9d/tutorial_dnn_yolo.html
https://blog.xugaoxiang.com/ai/darknet.html

在OpenCV中使用YOLO v3进行物体检测相关推荐

OpenCV中使用YOLO对象检测
点击上方"小白学视觉",选择加"星标"或"置顶" 重磅干货,第一时间送达本文转自:opencv学堂 OpenCV在3.3.1的版本中开始正 ...
yolo v3做行人检测+alexnet做上衣颜色识别
参考链接: 1)alexnet做上衣颜色识别 2)keras训练自己的数据集前言: 本项目是基于yolo v3已经训练好的coco数据集的模型进行行人检测的,然后将检测出来的结果传入到alexnet ...
OpenCV中的几种角点检测方法
1.Harris角点检测 import numpy as np import cv2 as cv import matplotlib.pyplot as pltplt.rcParams['font.s ...
如何在OpenCV中使用YOLO
点击上方"小白学视觉",选择加"星标"或"置顶" 重磅干货,第一时间送达今天,我们将研究如何在OpenCV框架中使用YOLO.YOLO于2 ...
目标检测 YOLO v3 训练人脸检测模型
YOLO,是You Only Look Once的缩写,一种基于深度卷积神经网络的物体检测算法,YOLO v3是YOLO的第3个版本,检测算法更快更准. 本文源码:https://github.com ...
.pth转.weights/openCV-python + YOLO v3实现目标检测
文章目录前言一.基于Pytorch框架的YOLO v3 二.openCV-python 三..pth转.weights 四模型部署总结前言毕设做的是水面目标的目标检测,因为要用Tkin ...
OpenCV中基于LBP算法的人脸检测测试代码
下面是OpenCV 3.3中基于CascadeClassifier类的LBP算法实现的人脸检测,从结果上看,不如其它开源库效果好,如libfacedetection,可参考 https://blog. ...
TensorFlow + Keras 实战 YOLO v3 目标检测图文并茂教程
作者 | fendouai 编辑 | 安可 [导读]:本文介绍了目标检测算法yolov3的keras实战.欢迎大家点击上方蓝字关注我们的公众号:深度学习与计算机视觉. YOLO 是一种非常流行的目标检 ...
探索 YOLO v3 实现细节 - 第6篇预测 (完结)
YOLO,即You Only Look Once的缩写,是一个基于卷积神经网络(CNN)的物体检测算法.而YOLO v3是YOLO的第3个版本,即YOLO.YOLO 9000.YOLO v3,检测效果 ...
对象检测目标小用什么模型好_自动驾驶目标检测- YOLO v3 深入解析
从2016年 Joseph Redmon 发布第一代YOLO开始,YOLO已经更新四代了,凭借着在一个网络模型中完成对图像中所有对象边界框和类别预测的独特创新,成为现今使用最广泛也是最快的对象检测算法 ...

在OpenCV中使用YOLO v3进行物体检测