yolov3聚类自己数据的anchor box

前言

yolov3是一个很优秀的object-detection模型，其中的anchor box机制在多尺度检测上取得了不错的效果。然而，作者提供的anchor box值是基于voc和coco数据集上的，如果应用到自己数据集可能不完全适用，那么如何基于自己的训练数据聚类anchor box呢？好吧，源代码如下所示。

kemans.py

import numpy as np


def iou(box, clusters):
    """
    Calculates the Intersection over Union (IoU) between a box and k clusters.
    :param box: tuple or array, shifted to the origin (i. e. width and height)
    :param 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)

    return iou_


def avg_iou(boxes, clusters):
    """
    Calculates the average Intersection over Union (IoU) between a numpy array of boxes and k clusters.
    :param boxes: numpy array of shape (r, 2), where r is the number of rows
    :param 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])])


def translate_boxes(boxes):
    """
    Translates all the boxes to the origin.
    :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)


def kmeans(boxes, k, dist=np.median):
    """
    Calculates k-means clustering with the Intersection over Union (IoU) metric.
    :param boxes: numpy array of shape (r, 2), where r is the number of rows
    :param k: number of clusters
    :param 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)]

    while True:
        for row in range(rows):
            distances[row] = 1 - iou(boxes[row], clusters)

        nearest_clusters = np.argmin(distances, axis=1)

        if (last_clusters == nearest_clusters).all():
            break

        for cluster in range(k):
            clusters[cluster] = dist(boxes[nearest_clusters == cluster], axis=0)

        last_clusters = nearest_clusters

    return clusters

cluster.py

import glob
import xml.etree.ElementTree as ET
import numpy as np
from kemans import kmeans, avg_iou

ANNOTATIONS_PATH = "F:/garbage/annotations"
CLUSTERS = 12


def load_dataset(path):
    dataset = []
    for xml_file in glob.glob("{}/*xml".format(path)):
        tree = ET.parse(xml_file)

        height = int(tree.findtext("./size/height"))
        width = int(tree.findtext("./size/width"))

        for obj in tree.iter("object"):
            xmin = int(obj.findtext("bndbox/xmin")) / width
            ymin = int(obj.findtext("bndbox/ymin")) / height
            xmax = int(obj.findtext("bndbox/xmax")) / width
            ymax = int(obj.findtext("bndbox/ymax")) / height

            xmin = np.float64(xmin)
            ymin = np.float64(ymin)
            xmax = np.float64(xmax)
            ymax = np.float64(ymax)
            if xmax == xmin or ymax == ymin:
                print(xml_file)
            dataset.append([xmax - xmin, ymax - ymin])
    return np.array(dataset)


if __name__ == '__main__':
    # print(__file__)
    data = load_dataset(ANNOTATIONS_PATH)
    out = kmeans(data, k=CLUSTERS)
    # clusters = [[10,13],[16,30],[33,23],[30,61],[62,45],[59,119],[116,90],[156,198],[373,326]]
    # out= np.array(clusters)/416.0
    print(out)
    print("Accuracy: {:.2f}%".format(avg_iou(data, out) * 100))
    print("Boxes:\n {}-{}".format(out[:, 0] * 608, out[:, 1] * 608))

    ratios = np.around(out[:, 0] / out[:, 1], decimals=2).tolist()
    print("Ratios:\n {}".format(sorted(ratios)))

最终聚类的结果：

[[0.3770724  0.15486111]
 [0.03958333 0.0375    ]
 [0.20677083 0.13686314]
 [0.0703125  0.05347222]
 [0.17916667 0.08472222]
 [0.47375    0.24791667]
 [0.08489583 0.08819444]
 [0.125      0.0625    ]
 [0.12447917 0.11319444]
 [0.28541667 0.1025    ]
 [0.2640625  0.21597222]
 [0.14942708 0.1875    ]]
Accuracy: 75.99%
Boxes:
 [229.26001999  24.06666667 125.71666667  42.75       108.93333333
 288.04        51.61666667  76.          75.68333333 173.53333333
 160.55        90.85166667]-[ 94.15555556  22.8         83.21278721  32.51111111  51.51111111
 150.73333333  53.62222222  38.          68.82222222  62.32
 131.31111111 114.        ]
Ratios:
 [0.8, 0.96, 1.06, 1.1, 1.22, 1.31, 1.51, 1.91, 2.0, 2.11, 2.43, 2.78]