ICode9

精准搜索请尝试: 精确搜索
首页 > 其他分享> 文章详细

YOLOV5——使用 k-means 聚类 anchorbox 数据

2021-01-08 20:03:13  阅读:1299  来源: 互联网

标签:YOLOV5 means iou shape boxes anchorbox 聚类 np clusters


训练的标注数据格式如下:

[
    {
        "name": "235_2_t20201127123021723_CAM2.jpg",
        "image_height": 6000,
        "image_width": 8192,
        "category": 5,
        "bbox": [
            1876.06,
            998.04,
            1883.06,
            1004.04
        ]
    },
    {
        "name": "235_2_t20201127123021723_CAM2.jpg",
        "image_height": 6000,
        "image_width": 8192,
        "category": 5,
        "bbox": [
            1655.06,
            1094.04,
            1663.06,
            1102.04
        ]
    }
]

聚类anchorbox只需要 bbox 中的左上角与右下角的 x,y 数据

 

k-means 聚类代码:

import numpy as np
import json


def iou(box, clusters):
    """
   计算 IOU
    param:
        box: tuple or array, shifted to the origin (i. e. width and height)
        clusters: numpy array of shape (k, 2) where k is the number of clusters
    return:
        numpy array of shape (k, 0) where k is the number of clusters
    """
    x = np.minimum(clusters[:, 0], box[0])
    y = np.minimum(clusters[:, 1], box[1])
    if np.count_nonzero(x == 0) > 0 or np.count_nonzero(y == 0) > 0:
        raise ValueError("Box has no area")

    intersection = x * y
    box_area = box[0] * box[1]
    cluster_area = clusters[:, 0] * clusters[:, 1]

    iou_ = intersection / (box_area + cluster_area - intersection + 1e-10)

    return iou_


#  计算框的 numpy 数组和 k 个簇之间的平均并集交集(IoU)。
def avg_iou(boxes, clusters):
    """
    param:
        boxes: numpy array of shape (r, 2), where r is the number of rows
        clusters: numpy array of shape (k, 2) where k is the number of clusters
    return:
        average IoU as a single float
    """
    return np.mean([np.max(iou(boxes[i], clusters)) for i in range(boxes.shape[0])])


# 将所有框转换为原点。
def translate_boxes(boxes):
    """
    param:
        boxes: numpy array of shape (r, 4)
    return:
    numpy array of shape (r, 2)
    """
    new_boxes = boxes.copy()
    for row in range(new_boxes.shape[0]):
        new_boxes[row][2] = np.abs(new_boxes[row][2] - new_boxes[row][0])
        new_boxes[row][3] = np.abs(new_boxes[row][3] - new_boxes[row][1])
    return np.delete(new_boxes, [0, 1], axis=1)


# 使用联合上的交集(IoU)度量计算k均值聚类。
def kmeans(boxes, k, dist=np.median):
    """
    param:
        boxes: numpy array of shape (r, 2), where r is the number of rows
        k: number of clusters
        dist: distance function
    return:
        numpy array of shape (k, 2)
    """
    rows = boxes.shape[0]

    distances = np.empty((rows, k))
    last_clusters = np.zeros((rows,))

    np.random.seed()

    # the Forgy method will fail if the whole array contains the same rows
    clusters = boxes[np.random.choice(rows, k, replace=False)]  # 初始化k个聚类中心(方法是从原始数据集中随机选k个)

    while True:
        for row in range(rows):
            # 定义的距离度量公式:d(box,centroid)=1-IOU(box,centroid)。到聚类中心的距离越小越好,但IOU值是越大越好,所以使用 1 - IOU,这样就保证距离越小,IOU值越大。
            distances[row] = 1 - iou(boxes[row], clusters)
        # 将标注框分配给“距离”最近的聚类中心(也就是这里代码就是选出(对于每一个box)距离最小的那个聚类中心)。
        nearest_clusters = np.argmin(distances, axis=1)
        # 直到聚类中心改变量为0(也就是聚类中心不变了)。
        if (last_clusters == nearest_clusters).all():
            break
        # 更新聚类中心(这里把每一个类的中位数作为新的聚类中心)
        for cluster in range(k):
            clusters[cluster] = dist(boxes[nearest_clusters == cluster], axis=0)

        last_clusters = nearest_clusters

    return clusters


# 读取 json 文件中的标注数据
def parse_anno(annotation_path):
    with open(annotation_path, 'r') as f:
        anno = json.load(f)
    result = []
    for line in anno:
        bbox = line['bbox']
        x_min, y_min, x_max, y_max = bbox[0], bbox[1], bbox[2], bbox[3]
        # 计算边框的大小
        width = x_max - x_min
        height = y_max - y_min
        assert width > 0
        assert height > 0
        result.append([width, height])
    result = np.asarray(result)
    return result


def get_kmeans(anno, cluster_num=9):

    anchors = kmeans(anno, cluster_num)
    ave_iou = avg_iou(anno, anchors)

    anchors = anchors.astype('int').tolist()

    anchors = sorted(anchors, key=lambda x: x[0] * x[1])

    return anchors, ave_iou


if __name__ == '__main__':
    annotation_path = "tile_round1_train_20201231/train_annos.json"
    anno_result = parse_anno(annotation_path)

    anchors, ave_iou = get_kmeans(anno_result, 9)

    anchor_string = ''
    for anchor in anchors:
        anchor_string += '{},{}, '.format(anchor[0], anchor[1])
    anchor_string = anchor_string[:-2]
   print(f'anchors are: {anchor_string}')
   print(f'the average iou is: {ave_iou}')


 

 

每次运行的结果都会有点不大一样

参考:https://blog.csdn.net/zuliang001/article/details/90551798

标签:YOLOV5,means,iou,shape,boxes,anchorbox,聚类,np,clusters
来源: https://www.cnblogs.com/yxyun/p/14253168.html

本站声明: 1. iCode9 技术分享网(下文简称本站)提供的所有内容,仅供技术学习、探讨和分享;
2. 关于本站的所有留言、评论、转载及引用,纯属内容发起人的个人观点,与本站观点和立场无关;
3. 关于本站的所有言论和文字,纯属内容发起人的个人观点,与本站观点和立场无关;
4. 本站文章均是网友提供,不完全保证技术分享内容的完整性、准确性、时效性、风险性和版权归属;如您发现该文章侵犯了您的权益,可联系我们第一时间进行删除;
5. 本站为非盈利性的个人网站,所有内容不会用来进行牟利,也不会利用任何形式的广告来间接获益,纯粹是为了广大技术爱好者提供技术内容和技术思想的分享性交流网站。

专注分享技术,共同学习,共同进步。侵权联系[81616952@qq.com]

Copyright (C)ICode9.com, All Rights Reserved.

ICode9版权所有