目标检测------锚框-------anchor box

锚框：以每一个像素点为中心生成多个大小和宽高比不同的边界框

图像 高h , 宽w

大小为s∈(0,1]  且   宽高比为r>0

锚框的高 h1 = hs/√r

锚框的宽 w1 = ws√r

s1，...，sn 与 r1，... ,rm  这样一张图的锚框有 whnm 个

但对于一点来说，我们感兴趣的框有：(s1,r1),(s1,r2)......(s1,rm),(s2,r1),......(sn,r1)

假如  w=416  h=416  s=[0.75,0.5,0.25]  r=[1,2,,0.5]   在 （200，200 ）处生成5个框：

[44, 44, 356, 356], [96, 96, 304, 304], [148, 148, 252, 252], [0, 90, 416, 310], [90, 0, 310, 416]

accard系数（Jaccard index）可以衡量两个集合的相似度。给定集合A和B，它们的Jaccard系数即二者交集大小除以二者并集大小：

J(A,B)=|A∩B|/|A∪B|

import math
import numpy as np

w = 416
h = 416
s = [0.75, 0.5, 0.25]
r = [1, 2, 0.5]


def create_box(x, y,imgw, imgh, sizes, ratios):
    # nn = int(len(sizes) + len(ratios) - 1)
    size_1 = sizes[0]
    ratio_1 = ratios[0]
    ratio_2 = ratios[1:]
    x = x / imgw
    y = y / imgh
    list_1 = [[
        x - i * math.sqrt(ratio_1) / 2,
        y - i / (2 * math.sqrt(ratio_1)),
        x + i * math.sqrt(ratio_1) / 2,
        y + i / (2 * math.sqrt(ratio_1))
    ] for i in sizes]
    list_2 = [[
        x - size_1 * math.sqrt(i) / 2,
        y - size_1 / (2 * math.sqrt(i)),
        x + size_1 * math.sqrt(i) / 2,
        y + size_1 / (2 * math.sqrt(i))
    ] for i in ratio_2]
    list_1.extend(list_2)
    list_all = []
    for l in list_1:
        if l[0] < 0:
            l[0] = 0
        else:
            l[0] = round(l[0] * imgw)

        if l[1] < 0:
            l[1] = 0
        else:
            l[1] = round(l[1] * imgh)

        if l[2] > 1:
            l[2] = imgw
        else:
            l[2] = round(l[2] * imgw)

        if l[3] > 1:
            l[3] = imgh
        else:
            l[3] = round(l[3] * imgh)
        list_all.append(l)
    return list_all
if __name__ == '__main__':
    print(create_box(200, 200, w, h, s, r))

　　

 

 

 

 

转载于:https://www.cnblogs.com/coylee/p/10887244.html