图像特征与目标检测1-HOG

迷南。 2023-01-10 15:55 171阅读 0赞

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dpbmdneW4_size_16_color_FFFFFF_t_70][]

HOG实现过程：

1灰度化（将图像看做一个x,y,z(灰度）的三维图像）

2采用Gamma校正法对输入图像进行颜色空间的标准化（归一化）

3计算图像每个像素的梯度（包括大小和方向）

4将图像划分为cells

5统计每个cell的梯度直方图（不同梯度的个数），得到cell的描述子

6将每几个cell组成一个block，得到block的描述子

7将图像image内的所有block的HOG特征descriptor串联起来就可以得到HOG特征，该特征向量就是用来目标检测或分类的特征。

import cv2
    import numpy as np
    #判断矩形i是否完全包含在矩形O中
    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
    #对人体绘制颜色框
    def draw_person(image,person):
        x,y,w,h = person
        cv2.rectangle(image,(x,y),(x+w,y+h),(0,255,255),2)
    img = cv2.imread("test5.jpg")
    hog = cv2.HOGDescriptor()
    hog.setSVMDetector(cv2.HOGDescriptor_getDefaultPeopleDetector())
    found,w = hog.detectMultiScale(img,0.1,(1,1))
    print(found)
    print(w)
    #丢弃某些完全被其他矩形包含在内的矩形
    found_filtered=[]
    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)
            print(found_filtered)
    #对不包含在内的有效矩形进行颜色框定
    for person in found_filtered:
        draw_person(img,person)
    cv2.imshow("people detection",img)
    cv2.waitKey(0)
    cv2.destroyAllWindows()

结果：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dpbmdneW4_size_16_color_FFFFFF_t_70 1][]

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dpbmdneW4_size_16_color_FFFFFF_t_70 2][]

原图来源：百度图库

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dpbmdneW4_size_16_color_FFFFFF_t_70 3][]

步长为(1,1)时候：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dpbmdneW4_size_16_color_FFFFFF_t_70 4][]

代码中由 ![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dpbmdneW4_size_16_color_FFFFFF_t_70 5][]

(1,1)->(2,2)

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dpbmdneW4_size_16_color_FFFFFF_t_70 6][]

改成（3，3）结果如下：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dpbmdneW4_size_16_color_FFFFFF_t_70 7][]

改成（7，7），据说。。最多检测5人。

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dpbmdneW4_size_16_color_FFFFFF_t_70 8][]

[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dpbmdneW4_size_16_color_FFFFFF_t_70]: /images/20221119/1892cbc6d46c49af9d469a8f969cdeb0.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dpbmdneW4_size_16_color_FFFFFF_t_70 1]: https://img-blog.csdnimg.cn/20210125150048790.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dpbmdneW4=,size_16,color_FFFFFF,t_70
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dpbmdneW4_size_16_color_FFFFFF_t_70 2]: https://img-blog.csdnimg.cn/20210125145852137.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dpbmdneW4=,size_16,color_FFFFFF,t_70
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dpbmdneW4_size_16_color_FFFFFF_t_70 3]: https://img-blog.csdnimg.cn/20210125150614778.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dpbmdneW4=,size_16,color_FFFFFF,t_70
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dpbmdneW4_size_16_color_FFFFFF_t_70 4]: https://img-blog.csdnimg.cn/20210125150821403.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dpbmdneW4=,size_16,color_FFFFFF,t_70
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dpbmdneW4_size_16_color_FFFFFF_t_70 5]: https://img-blog.csdnimg.cn/2021012515092715.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dpbmdneW4=,size_16,color_FFFFFF,t_70
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dpbmdneW4_size_16_color_FFFFFF_t_70 6]: https://img-blog.csdnimg.cn/20210125151435718.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dpbmdneW4=,size_16,color_FFFFFF,t_70
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dpbmdneW4_size_16_color_FFFFFF_t_70 7]: https://img-blog.csdnimg.cn/20210125151306100.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dpbmdneW4=,size_16,color_FFFFFF,t_70
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dpbmdneW4_size_16_color_FFFFFF_t_70 8]: https://img-blog.csdnimg.cn/20210125151702604.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dpbmdneW4=,size_16,color_FFFFFF,t_70