python-opencv3计算机视觉学习笔记（七）

七：目标检测与识别

梯度直方图（Histogram of Oriented Gradient）

图像金字塔（image pyramid）

滑动窗口（sliding window）

7.1、目标检测与识别

HOG描述符（详情见opencv特征提取描述符）

将图像分成小单元（16*16），每个单元包含视觉表示，安八个方向（N,NW,W,SW,W,SW,S,SE,E,NE）计算颜色梯度

尺度：？？

位置：检测目标可能位于图像任何地方，需扫描图像各个部分，找出感兴趣区域，并尝试检测目标

为解决位置、尺寸问题需熟悉图像金字塔和滑动窗口概念

非最大（极大)抑制：对图像同一区域相关的结果进行抑制，只关心结果最好的窗口，丢弃评分低的重叠窗口。

支持向量机SVM：

对于带有标签的训练数据，通过一个优化的超平面来对这些数据进行分类

2、检测人（opencv的HOG）

#!/usr/bin/python

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

# @Time : 2017/06/10

# @Author :

# @Site : 检测人

# @Software: PyCharm

#检测行人
import cv2
import numpy as np

#如果矩形被完全包含在另外一个矩形中，可确定该矩形应该被丢弃
def is_inside(o, i):
    ox, oy, ow, oh = o
    ix, iy, iw, ih = i
    return ox > ix and oy > iy and ox + ow <ix + iw and oy +oh < iy +ih

def draw_person(image, person):
    x, y, w, h = person
    # cv2.rectangle(img, (x, y), (x+w, y+h), (0,255,255), 2)
    cv2.rectangle(image, (x, y), (x + w, y + h), (0, 0, 255), 2)

    # person = np.array([[x, y, x + w, y + h] for (x, y, w, h) in person])
    # pick = non_max_suppression(person, probs=None, overlapThresh=0.65)

img = cv2.imread("C:\\Software\\Python\\snapshotoutdoor\\rightRGB_11.jpg")
hog = cv2.HOGDescriptor()  # 检测人的默认检测器 内置目标检测器 实际效果并不好（已经训练好的模型？？）
hog.setSVMDetector(cv2.HOGDescriptor_getDefaultPeopleDetector())  # SVM

found, w = hog.detectMultiScale(img)  # 加载图像

found_filtered = []
# 遍历检测结果来丢弃不含有检测目标的区域
for ri, r in enumerate(found):
    for qi, q in enumerate(found):
        if ri != qi and is_inside(r, q):
            break
        else:
            found_filtered.append(r) #append 什么作用？？

for person in found_filtered:
    draw_person(img, person)

cv2.imshow("people detection", img)


cv2.waitKey(0)
cv2.destroyAllWindows()

另：上述方法对近景人检测效果不佳，下面来自一老外的方法，近景效果不错

#检测近景的效果与test_ObjectDetection.py效果好
# import the necessary packages
from __future__ import print_function
from imutils.object_detection import non_max_suppression
from imutils import paths
import numpy as np
import argparse
import imutils
import cv2

# left_camera1 = cv2.VideoCapture(0)

# construct the argument parse and parse the arguments
# ap = argparse.ArgumentParser()# the path to the directory that contains the list of images we are going to perform pedestrian detection on.
# ap.add_argument("-i", "--images", required=True, help="path to images directory")
# args = vars(ap.parse_args())
imagePath = "C:\\Software\\Python\\snapshotoutdoor\\rightRGB_11.jpg"
# initialize the HOG descriptor/person detector
hog = cv2.HOGDescriptor()
hog.setSVMDetector(cv2.HOGDescriptor_getDefaultPeopleDetector())
# loop over the image paths
# for imagePath in paths.list_images(args["images"]):#start looping over the images in our --images  directory
    # load the image and resize it to (1) reduce detection time
    # and (2) improve detection accuracy
while True:
    # ret1, frame1 = left_camera1.read()
    # cv2.imshow("frame1", frame1)
    # cv2.waitKey(10)
    # image = frame1
    image = cv2.imread(imagePath)

    # image = imutils.resize(image, width=min(400, image.shape[1]))
    orig = image.copy()

    # detect people in the image
    (rects, weights) = hog.detectMultiScale(image, winStride=(4, 4),
                                            padding=(8, 8), scale=1.05)

    # draw the original bounding boxes
    for (x, y, w, h) in rects:
        cv2.rectangle(orig, (x, y), (x + w, y + h), (0, 0, 255), 2)

    # apply non-maxima suppression to the bounding boxes using a
    # fairly large overlap threshold to try to maintain overlapping
    # boxes that are still people
    rects = np.array([[x, y, x + w, y + h] for (x, y, w, h) in rects])
    pick = non_max_suppression(rects, probs=None, overlapThresh=0.65)

    # draw the final bounding boxes
    for (xA, yA, xB, yB) in pick:
        cv2.rectangle(image, (xA, yA), (xB, yB), (0, 255, 0), 2)

    # show some information on the number of bounding boxes
    # filename = imagePath[imagePath.rfind("/") + 1:]
    # print("[INFO] {}: {} original boxes, {} after suppression".format(
    #     filename, len(rects), len(pick)))

    # show the output images
    cv2.imshow("Before NMS", orig)
    cv2.imshow("After NMS", image)
    cv2.waitKey(1)

3、创建和训练目标检测器

使用SVM和词袋（Bag of Word,BOW）

词袋：计算文档中每个词出现的次数，用该些次数构成的向量重新表示文档

示列如下:

I like opencv and i like python

I like c++ and python

I don’t like artichokes

可以用以下值来建立字

{

I:4,

Like:4,

Opencv:2,

And:2

Python:2

C++:1

Don’t:1

Artichokes:1

}

以上的三句话可以用以下向量表示:

[2,2,1,1,1,0,0,0]

[1,1,0,1,1,1,0,0]

[1,1,0,0,0,0,1,1]

代码实现？？？以后每学一个知识要留下学习过程并整理保存https://blog.csdn.net/lilai619/article/details/46740837

BoW(词袋模型)+python代码实现

？？？？

K-means聚类：数据分析的向量量化方法

？？？

python-opencv3计算机视觉学习笔记（七）

猜你喜欢