Object detection with deep learning and OpenCV

时间：2024-09-19 17:35:14

  这篇文章只是基于OpenCV使用SSD算法执行目标检测；不涉及到SSD的理论原理、不涉及训练过程；也就是说仅仅使用训练好的模型文件基于OpenCV做测试；包括图片和视频；

  只用作笔记，原教程地址：Object detection with deep learning and OpenCV

Single Shot Detectors for Object Detection

  当提到基于深度学习的目标检测算法，大家都多多少少的听说过这三种算法：

 当然了，现在已经是19年了，上面三种算法也已经更新换代了；那之所以还列举出来，想要表达的是这三类算法是相当good,...(完了，装不下去了....)

R-CNN系列检测算法，精确度高，速度慢；

YOLO系列检测算法，速度快，精确度有些欠缺；

SSD取了两者的优点吧。。。。

Deep learning-based object detection with OpenCV

Object detection with deep learning and OpenCV

#!/usr/bin/env python

#-*- coding:utf-8 -*-

# @Time    : 19-4-24 下午3:52

# @Author  : chen

"""

利用MobileNet SSD + OpenCV中的dnn执行目标检测

python deep_learning_object_detection_cz.py --image images/face_1.jpg \

--prototxt MobileNetSSD_deploy.prototxt.txt \

 --model MobileNetSSD_deploy.caffemodel

"""

# 依赖包

import numpy as np

import argparse

import cv2

import time

import pdb

# 解析命令行参数

ap = argparse.ArgumentParser()

ap.add_argument("-i", "--image", required=True, help="path to input image")

ap.add_argument("-p", "--prototxt", required=True, help="path to Caffe 'deploy' prototxt file")

ap.add_argument("-m", "--model", required=True, help="path to Caffe pre-trained model")

ap.add_argument("-c", "--confidence", type=float, default=0.2, help="minimum probability to filter weak detections")

args = vars(ap.parse_args())

# 初始化类标签，然后为每一个类别设置一个颜色值

# 该颜色值是为了在图像中画出矩形框的时候使用

CLASSES = ["background", "aeroplane", "bicycle", "bird", "boat",

           "bottle", "bus", "car", "cat", "chair", "cow", "diningtable",

           "dog", "horse", "motorbike", "person", "pottedplant", "sheep",

           "sofa", "train", "tvmonitor"]

COLORS = np.random.uniform(0, 255, size=(len(CLASSES), 3))

# 加载训练好的Caffe模型

# OpenCV的dnn方法中，可以加载由Caffe，TensorFLow，Darknet，Torch训练得到的模型文件的方法

print("[INFO] loading model...")

net = cv2.dnn.readNetFromCaffe(args["prototxt"], args["model"])

# 加载测试图片，并转换成blob(OpenCV需要这样做)

image = cv2.imread(args["image"])

(h, w) = image.shape[:2]

# cv2.dnn.blobFromImage返回一个四维的blob

# 可以对image执行缩放，剪切，交换RB通道，减均值操作

blob = cv2.dnn.blobFromImage(cv2.resize(image, (300, 300)), 0.007843, (300, 300), 127.5)

# 输入到网络中，执行Inference

print("[INFO] computing object detections...")

net.setInput(blob)

start = time.time()

detections = net.forward()

end = time.time()

print("[INFO] SSD took {:6} seconds.".format(end - start))

# pdb.set_trace()

for i in range(detections.shape[2]):

    # 类别概率

    confidence = detections[0, 0, i, 2]

    # 过滤掉confidence小于人为设定的阈值的detection

    if confidence > args["confidence"]:

        idx = int(detections[0, 0, i, 1])  # 类别索引

        box = detections[0, 0, i, 3:7] * np.array([w, h, w, h])

        (startX, startY, endX, endY) = box.astype("int")  # SSD的输出直接就是框的左上角和右下角的点的坐标位置

        # 在图片展示检测的object

        label = "{}: {:.2f}%".format(CLASSES[idx], confidence*100)

        print("[INFO] {}".format(label))

        cv2.rectangle(image, (startX, startY), (endX, endY), COLORS[idx], 2)

        y = startY -15 if startY - 15 > 15 else startY + 15

        cv2.putText(image, label, (startX, y), cv2.FONT_HERSHEY_SIMPLEX, 0.5, COLORS[idx], 2)

# 显示

cv2.imshow("Object Detection", image)

cv2.waitKey(0)

cv2.imwrite("lab_2_ssd.jpg", image)

 这样有没有用？？？？？用处不大

  还是需要看论文的。。。。。

相关文章