python开发：基于SSD下的图像内容识别（一）1.首先，先看一下我们能达到什么样的效果：2.我们需要做哪些基本的步骤：

import  cv2
import selectivesearch
import matplotlib.pyplot as  plt
import matplotlib.patches as  mpatches
import  numpy as  np
img = cv2.imread('/Users/slade/Documents/Yoho/picture_recognize/heshen.jpg')

#图片识别分割
img_lbl, regions =selectivesearch.selective_search(
    img, scale=500, sigma=0.9, min_size=20)

#这边的regions里面就有一个个划分出来的box
regions[0]
#Out[3]: {'labels': [0.0], 'rect': (0, 0, 619, 620), 'size': 177325},其中‘rect’定位了box的位置，‘size’确定了box的大小

# 接下来我们把窗口和图像打印出来，对它有个直观认识
fig, ax = plt.subplots(ncols=1, nrows=1, figsize=(6, 6))
ax.imshow(img)

for reg in regions:
    x, y, w, h = reg['rect']
    rect = mpatches.Rectangle(
        (x, y), w, h, fill=False, edgecolor='red', linewidth=1)
    ax.add_patch(rect)
plt.show()

candidates = []
for r in regions:
    # 重复的不要
    if r['rect'] in candidates:
        continue
    # 太小和太大的不要
    if r['size'] < 200 or r['size']>20000:
        continue

    x, y, w, h = r['rect']
    # 太不方的不要
    if w / h > 1.8 or h / w > 1.8:
        continue
    candidates.append((x,y,w,h))

#剔除大box内的小box
candidates_sec = []
for  i  in  candidates:
    if len(candidates_sec)==0:
        candidates_sec.append(i)
    else:
        Flag=False
        replace=-1
        index=0
        for j  in candidates_sec:
            ##新box在小圈 则删除
            if i[0]>=j[0] and i[0]+i[2]<=j[0]+j[2]  and i[1]>=j[1] and i[1]+i[3]<=j[1]+j[3]: 
                Flag=True
                break
            ##新box不在小圈 而在老box外部 替换老box
            elif i[0]<=j[0] and i[0]+i[2]>=j[0]+j[2] and i[1]<=j[1] and i[1]+i[3]>=j[1]+j[3]: 
                replace=index
                break
            index+=1

        if not Flag:
            if replace>=0:
                candidates_sec[replace]=i
            else:
                candidates_sec.append(i)

fig, ax = plt.subplots(ncols=1, nrows=1, figsize=(6, 6))
ax.imshow(img)
for x, y, w, h in candidates_sec:
    rect = mpatches.Rectangle(
        (x, y), w, h, fill=False, edgecolor='red', linewidth=1)
    ax.add_patch(rect)
plt.show()

def sim_fill(r1, r2, imsize):
    """
        calculate the fill_similarity over the image
    """
    bbsize = (
        (max(r1["max_x"], r2["max_x"]) - min(r1["min_x"], r2["min_x"]))
        * (max(r1["max_y"], r2["max_y"]) - min(r1["min_y"], r2["min_y"]))
    )
    return 1.0 - (bbsize - r1["size"] - r2["size"]) / imsize