非极大抑制（Non-maximum suppression）python代码实现

定位一个物体，最后算法就找出了一堆的方框，我们需要判别哪些矩形框是没用的。非极大值抑制：先假设有6个矩形框，根据分类器类别分类概率做排序，从大到小分别属于物体的概率分别为A、B、C、D、E、F。

(1)从最大概率矩形框F开始，分别判断B~F与A的重叠度IOU是否大于某个设定的阈值;

(2)假设B、D与F的重叠度超过阈值，那么就扔掉B、D；并标记第一个矩形框A，是我们保留下来的。

(3)从剩下的矩形框C、E、F中，选择概率最大的C，然后判断C与E、F的重叠度，重叠度大于一定的阈值，那么就扔掉；并标记C是我们保留下来的第二个矩形框。

就这样一直重复，找到所有被保留下来的矩形框。
 

# import the necessary packages
import numpy as np


#  Felzenszwalb et al.
def non_max_suppression_slow(boxes, thresh):
    # if there are no boxes, return an empty list
    if len(boxes) == 0:
        return []

    # initialize the list of picked indexes
    pick = []

    # grab the coordinates of the bounding boxes
    x1 = boxes[:, 0]
    y1 = boxes[:, 1]
    x2 = boxes[:, 2]
    y2 = boxes[:, 3]
    scores = boxes[:, 4]

    # compute the area of the bounding boxes and sort the bounding
    # boxes by the bottom-right y-coordinate of the bounding box
    area = (x2 - x1 + 1) * (y2 - y1 + 1)
    
    # 获取置信度，并降序排列，获取其在boxes中对应的索引
    idxs = scores.argsort()[::-1]

    while idxs.size > 0:
        i = idxs[0]
        pick.append(i)

        # find the largest (x, y) coordinates for the start of
        # the bounding box and the smallest (x, y) coordinates
        # for the end of the bounding box
        xx1 = np.maximum(x1[i], x1[idxs[1:]])
        yy1 = np.maximum(y1[i], y1[idxs[1:]])
        xx2 = np.minimum(x2[i], x2[idxs[1:]])
        yy2 = np.minimum(y2[i], y2[idxs[1:]])

        # compute the width and height of the bounding box
        w = np.maximum(0.0, xx2 - xx1 + 1)
        h = np.maximum(0.0, yy2 - yy1 + 1)

        # compute the intersection over union（IOU） between the computed
        # bounding box and the bounding box in the area list
        inter = w * h
        union = area[i] + area[idxs[1:]] - inter
        IOU = inter / union

        indexes = np.where(IOU < thresh)

        idxs = idxs[indexes + 1]

    # return only the bounding boxes that were picked
    return boxes[pick]

 

References:

1.https://www.pyimagesearch.com/2014/11/17/non-maximum-suppression-object-detection-python/

2.https://blog.csdn.net/l_ml_m_lm_m/article/details/79881437

3.https://blog.csdn.net/u011534057/article/details/51235718

4.http://www.cnblogs.com/makefile/p/nms.html