RCNN系列——Faster RCNN anchor生成解析_rpn生成初始anchor-CSDN博客

本文链接：https://blog.csdn.net/jmu201521121021/article/details/87877721

RCNN系列——Faster RCNN anchor生成解析

一、简介

Faster RCNN里的anchor 是一个矩形框，用于box regressor, 可以减少网络计算量，把anchor 数目定义为 $k$ ,featur map每个点只预测 $k$ 个box,然后与anchor计算偏差，回归更加精准的框位置。假设feature map 尺寸为 $\times H$ ,输出anchor总数目为 $\times H \times k$ .
Faster RCNN里面每个anchor都有 $x_{center},y_{center},w,h)$ 信息，也就是anchor的中心位置和长宽。还有 $s c a l e$ 和 $r a t i o$ 两个参数， $s c a l e = [8, 16, 32]$ , $r a t i o = [0.5, 1, 2]$ .scale是指anchor相对于基础anchor大小的比例，ratio是指面积相同的anchor 长宽比例,也就是说每一种scale的anchor都有3种面积相同,长宽不同的anchor，总共就有 $\times 3 = 9$ 个anchor。
后端网络是ZF时，基础anchor大小为 $16 \times 16$ ,也就是feature map 上一个点代表输入原图 $16 \times 16$ 区域。因为经过ZF网络提取的feature map的尺寸为原图的 $\frac{1}{16}$ ,所以基础大小为16。
经过scale 变换，anchor 大小有 $8\times16 = 128,16\times 16 = 256,32 \times 16 =512$ 三种尺寸。如下图，anchor起始坐标系为 $128\times 128$ 的左下角。
九个anchor的坐标信息 $（ x 1, y 1, x 2, y 2 ）$ (左上角和右下角)
anchors =
-83 -39 100 56
-175 -87 192 104
-359 -183 376 200
-55 -55 72 72
-119 -119 136 136
-247 -247 264 264
-35 -79 52 96
-79 -167 96 184
-167 -343 184 360

二、代码解析

generate_anchors.py
- 函数1：def generate_anchors(base_size=16, ratios=[0.5, 1, 2],scales=2**np.arange(3, 6))，功能：先生成base anchor $16 \times 16$ 大小的 $（ x 1, y 1, x 2, y 2 ）$ ，然后依次生成每个scale的ratio的anchor(3个)，最后生成9个。
- 函数2：_whctrs(anchor): 功能：求 $x_{center},y_{center},w,h）$
- 函数3：_ratio_enum(anchor, ratios)::功能：求所有ratios的anchor $x 1, y 1, x 2, y 2$
- 函数4：_mkanchors(ws, hs, x_ctr, y_ctr): 功能： $x_{center},y_{center},w,h）$ 转化为 $x 1, y 1, x 2, y 2$

# --------------------------------------------------------
# Faster R-CNN
# Copyright (c) 2015 Microsoft
# Licensed under The MIT License [see LICENSE for details]
# Written by Ross Girshick and Sean Bell
# --------------------------------------------------------

import numpy as np

# Verify that we compute the same anchors as Shaoqing's matlab implementation:
#
#    >> load output/rpn_cachedir/faster_rcnn_VOC2007_ZF_stage1_rpn/anchors.mat
#    >> anchors
#
#    anchors =
#
#       -83   -39   100    56
#      -175   -87   192   104
#      -359  -183   376   200
#       -55   -55    72    72
#      -119  -119   136   136
#      -247  -247   264   264
#       -35   -79    52    96
#       -79  -167    96   184
#      -167  -343   184   360

#array([[ -83.,  -39.,  100.,   56.],
#       [-175.,  -87.,  192.,  104.],
#       [-359., -183.,  376.,  200.],
#       [ -55.,  -55.,   72.,   72.],
#       [-119., -119.,  136.,  136.],
#       [-247., -247.,  264.,  264.],
#       [ -35.,  -79.,   52.,   96.],
#       [ -79., -167.,   96.,  184.],
#       [-167., -343.,  184.,  360.]])

def generate_anchors(base_size=16, ratios=[0.5, 1, 2],
                     scales=2**np.arange(3, 6)):
    """
    Generate anchor (reference) windows by enumerating aspect ratios X
    scales wrt a reference (0, 0, 15, 15) window.
    """
    base_anchor = np.array([1, 1, base_size, base_size]) - 1 
    ratio_anchors = _ratio_enum(base_anchor, ratios)
    anchors = np.vstack([_scale_enum(ratio_anchors[i, :], scales)
                         for i in xrange(ratio_anchors.shape[0])])
    return anchors

def _whctrs(anchor):
    """
    Return width, height, x center, and y center for an anchor (window).
    """

    w = anchor[2] - anchor[0] + 1
    h = anchor[3] - anchor[1] + 1
    x_ctr = anchor[0] + 0.5 * (w - 1)
    y_ctr = anchor[1] + 0.5 * (h - 1)
    return w, h, x_ctr, y_ctr

def _mkanchors(ws, hs, x_ctr, y_ctr):
    """
    Given a vector of widths (ws) and heights (hs) around a center
    (x_ctr, y_ctr), output a set of anchors (windows).
    """

    ws = ws[:, np.newaxis]
    hs = hs[:, np.newaxis]
    anchors = np.hstack((x_ctr - 0.5 * (ws - 1),
                         y_ctr - 0.5 * (hs - 1),
                         x_ctr + 0.5 * (ws - 1),
                         y_ctr + 0.5 * (hs - 1)))
    return anchors

def _ratio_enum(anchor, ratios):
    """
    Enumerate a set of anchors for each aspect ratio wrt an anchor.
    """

    w, h, x_ctr, y_ctr = _whctrs(anchor)
    size = w * h
    size_ratios = size / ratios
    ws = np.round(np.sqrt(size_ratios))
    hs = np.round(ws * ratios)
    anchors = _mkanchors(ws, hs, x_ctr, y_ctr)
    return anchors

def _scale_enum(anchor, scales):
    """
    Enumerate a set of anchors for each scale wrt an anchor.
    """

    w, h, x_ctr, y_ctr = _whctrs(anchor)
    ws = w * scales
    hs = h * scales
    anchors = _mkanchors(ws, hs, x_ctr, y_ctr)
    return anchors

if __name__ == '__main__':
    import time
    t = time.time()
    a = generate_anchors()
    print time.time() - t
    print a
    from IPython import embed; embed()