python物体标识怎么实现

这篇文章主要为大家展示了"python物体标识怎么实现"，内容简而易懂，条理清晰，希望能够帮助大家解决疑惑，下面让小编带领大家一起研究并学习一下"python物体标识怎么实现"这篇文章吧。

1、读取彩色图像进行灰度化和二值化。

def get_binary_img(img):    # gray img to bin image    bin_img = np.zeros(shape=(img.shape), dtype=np.uint8)    h = img.shape[0]    w = img.shape[1]    for i in range(h):        for j in range(w):            bin_img[i][j] = 255 if img[i][j] < 255 else 0    return bin_img# 调用file_name = "./test.bmp"img = cv2.imread(file_name)# 灰度化gray_img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)# 二值化bin_img = get_binary_img(gray_img)

2、目标标志，每个物体的像素值是该物体的标志，为计算面积打下基础。

# 标记目标def label_region(bin_img,width,height):    visited = np.zeros(shape=bin_img.shape,dtype=np.uint8)    label_img = np.zeros(shape=bin_img.shape, dtype=np.uint8)    label = 0    for i in range(height):        for j in range(width):            if bin_img[i][j] == 255 and visited[i][j]==0 : //找到种子点                # visit                visited[i][j] = 1                label += 1                label_img[i][j] = label                # label                label_from_seed(bin_img, visited, i, j, label, label_img)    return label_img# 区域增长法进行标记def label_from_seed(bin_img,visited,i,j,label,out_img):    directs = [(-1, -1), (0, -1), (1, -1), (1, 0), (1, 1), (0, 1), (-1, 1), (-1, 0)]    seeds = [(i,j)]    height = bin_img.shape[0]    width = bin_img.shape[1]    while len(seeds):        seed = seeds.pop(0)        i = seed[0]        j = seed[1]        if visited[i][j] == 0:            visited[i][j] = 1            out_img[i][j] = label         # 以（i,j）为起点进行标记        for direct in directs:            cur_i = i + direct[0]            cur_j = j + direct[1]             # 非法            if cur_i < 0 or cur_j < 0 or cur_i >= height or cur_j >= width:                continue             # 没有访问过            if visited[cur_i][cur_j] == 0 and bin_img[cur_i][cur_j] == 255:                visited[cur_i][cur_j] = 1                out_img[cur_i][cur_j] = label                seeds.append((cur_i,cur_j))

3、通过遍历标记的图像，统计每个编号中出现的像素数，可以得到不同区域的面积大小。

def get_region_area(label_img,label):    count = { key: 0  for key in range(label + 1)}    start_pt = {key:(0,0) for key in range(label + 1)}    height = label_img.shape[0]    width  = label_img.shape[1]    for i in range(height):        for j in range(width):            key = label_img[i][j]            count[key] += 1            if count[key] == 1:                start_pt[key] = (j,i)    return count,start_pt

以上是"python物体标识怎么实现"这篇文章的所有内容，感谢各位的阅读！相信大家都有了一定的了解，希望分享的内容对大家有所帮助，如果还想学习更多知识，欢迎关注行业资讯频道！