我们是如何改进YOLOv3进行红外小目标检测的？¶

#coding=utf-8
import xml.etree.ElementTree as ET
import numpy as np


def iou(box, clusters):
    """
    计算一个ground truth边界盒和k个先验框(Anchor)的交并比(IOU)值。
    参数box: 元组或者数据，代表ground truth的长宽。
    参数clusters: 形如(k,2)的numpy数组，其中k是聚类Anchor框的个数
    返回：ground truth和每个Anchor框的交并比。
    """
    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):
    """
    计算一个ground truth和k个Anchor的交并比的均值。
    """
    return np.mean([np.max(iou(boxes[i], clusters)) for i in range(boxes.shape[0])])

def kmeans(boxes, k, dist=np.median):
    """
    利用IOU值进行K-means聚类
    参数boxes: 形状为(r, 2)的ground truth框，其中r是ground truth的个数
    参数k: Anchor的个数
    参数dist: 距离函数
    返回值：形状为(k, 2)的k个Anchor框
    """
    # 即是上面提到的r
    rows = boxes.shape[0]
    # 距离数组，计算每个ground truth和k个Anchor的距离
    distances = np.empty((rows, k))
    # 上一次每个ground truth"距离"最近的Anchor索引
    last_clusters = np.zeros((rows,))
    # 设置随机数种子
    np.random.seed()

    # 初始化聚类中心，k个簇，从r个ground truth随机选k个
    clusters = boxes[np.random.choice(rows, k, replace=False)]
    # 开始聚类
    while True:
        # 计算每个ground truth和k个Anchor的距离，用1-IOU(box,anchor)来计算
        for row in range(rows):
            distances[row] = 1 - iou(boxes[row], clusters)
        # 对每个ground truth，选取距离最小的那个Anchor，并存下索引
        nearest_clusters = np.argmin(distances, axis=1)
        # 如果当前每个ground truth"距离"最近的Anchor索引和上一次一样，聚类结束
        if (last_clusters == nearest_clusters).all():
            break
        # 更新簇中心为簇里面所有的ground truth框的均值
        for cluster in range(k):
            clusters[cluster] = dist(boxes[nearest_clusters == cluster], axis=0)
        # 更新每个ground truth"距离"最近的Anchor索引
        last_clusters = nearest_clusters

    return clusters

# 加载自己的数据集，只需要所有labelimg标注出来的xml文件即可
def load_dataset(path):
    dataset = []
    for xml_file in glob.glob("{}/*xml".format(path)):
        tree = ET.parse(xml_file)
        # 图片高度
        height = int(tree.findtext("./size/height"))
        # 图片宽度
        width = int(tree.findtext("./size/width"))

        for obj in tree.iter("object"):
            # 偏移量
            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)
            # 将Anchor的长宽放入dateset，运行kmeans获得Anchor
            dataset.append([xmax - xmin, ymax - ymin])
    return np.array(dataset)

if __name__ == '__main__':

    ANNOTATIONS_PATH = "F:\Annotations" #xml文件所在文件夹
    CLUSTERS = 9 #聚类数量，anchor数量
    INPUTDIM = 416 #输入网络大小

    data = load_dataset(ANNOTATIONS_PATH)
    out = kmeans(data, k=CLUSTERS)
    print('Boxes:')
    print(np.array(out)*INPUTDIM)    
    print("Accuracy: {:.2f}%".format(avg_iou(data, out) * 100))       
    final_anchors = np.around(out[:, 0] / out[:, 1], decimals=2).tolist()
    print("Before Sort Ratios:\n {}".format(final_anchors))
    print("After Sort Ratios:\n {}".format(sorted(final_anchors)))

def kmean_anchors(path='./2007_train.txt', n=5, img_size=(416, 416)):
    # from utils.utils import *; _ = kmean_anchors()
    # Produces a list of target kmeans suitable for use in *.cfg files
    from utils.datasets import LoadImagesAndLabels
    thr = 0.20  # IoU threshold

    def print_results(thr, wh, k):
        k = k[np.argsort(k.prod(1))]  # sort small to large
        iou = wh_iou(torch.Tensor(wh), torch.Tensor(k))
        max_iou, min_iou = iou.max(1)[0], iou.min(1)[0]
        bpr, aat = (max_iou > thr).float().mean(), (
            iou > thr).float().mean() * n  # best possible recall, anch > thr
        print('%.2f iou_thr: %.3f best possible recall, %.2f anchors > thr' %
              (thr, bpr, aat))
        print(
            'kmeans anchors (n=%g, img_size=%s, IoU=%.3f/%.3f/%.3f-min/mean/best): '
            % (n, img_size, min_iou.mean(), iou.mean(), max_iou.mean()),
            end='')
        for i, x in enumerate(k):
            print('%i,%i' % (round(x[0]), round(x[1])),
                  end=',  ' if i < len(k) - 1 else '\n')  # use in *.cfg
        return k

    def fitness(thr, wh, k):  # mutation fitness
        iou = wh_iou(wh, torch.Tensor(k)).max(1)[0]  # max iou
        bpr = (iou > thr).float().mean()  # best possible recall
        return iou.mean() * bpr  # product

    # Get label wh
    wh = []
    dataset = LoadImagesAndLabels(path,
                                  augment=True,
                                  rect=True,
                                  cache_labels=True)
    nr = 1 if img_size[0] == img_size[1] else 10  # number augmentation repetitions
    for s, l in zip(dataset.shapes, dataset.labels):
        wh.append(l[:, 3:5] *
                  (s / s.max()))  # image normalized to letterbox normalized wh
    wh = np.concatenate(wh, 0).repeat(nr, axis=0)  # augment 10x
    wh *= np.random.uniform(img_size[0], img_size[1],
                            size=(wh.shape[0],
                                  1))  # normalized to pixels (multi-scale)

    # Darknet yolov3.cfg anchors
    use_darknet = False
    if use_darknet:
        k = np.array([[10, 13], [16, 30], [33, 23], [30, 61], [62, 45],
                      [59, 119], [116, 90], [156, 198], [373, 326]])
    else:
        # Kmeans calculation
        from scipy.cluster.vq import kmeans
        print('Running kmeans for %g anchors on %g points...' % (n, len(wh)))
        s = wh.std(0)  # sigmas for whitening
        k, dist = kmeans(wh / s, n, iter=30)  # points, mean distance
        k *= s
    k = print_results(thr, wh, k)
    # Evolve
    wh = torch.Tensor(wh)
    f, ng = fitness(thr, wh, k), 2000  # fitness, generations
    for _ in tqdm(range(ng), desc='Evolving anchors'):
        kg = (
            k.copy() *
            (1 + np.random.random() * np.random.randn(*k.shape) * 0.30)).clip(
                min=2.0)
        fg = fitness(thr, wh, kg)
        if fg > f:
            f, k = fg, kg.copy()
            print_results(thr, wh, k)
    k = print_results(thr, wh, k)

    return k

13, 18, 16, 22, 19, 25

12,17, 14,17, 15,19, 15,21, 13,20, 19,24

10,16, 12,17, 13,20, 13,22, 15,18, 15,20, 15,23, 18,23, 21,26

https://lutzroeder.github.io/netron/

import os
import shutil
cfg_path = "./cfg/yolov3-dwconv-cbam.cfg"
save_path = "./cfg/preprocess_cfg/"
new_save_name = os.path.join(save_path,os.path.basename(cfg_path))

f = open(cfg_path, 'r')
lines = f.readlines()

# 去除以#开头的，属于注释部分的内容
# lines = [x for x in lines if x and not x.startswith('#')]
# lines = [x.rstrip().lstrip() for x in lines]

lines_nums = []
layers_nums = []

layer_cnt = -1

for num, line in enumerate(lines):
    if line.startswith('['):
        layer_cnt += 1
        layers_nums.append(layer_cnt)
        lines_nums.append(num+layer_cnt)
        print(line)
        # s = s.join("")
    # s = s.join(line)
for i,num in enumerate(layers_nums):
    print(lines_nums[i], num)
    lines.insert(lines_nums[i]-1, '# layer-%d\n' % (num-1))
fo = open(new_save_name, 'w')
fo.write(''.join(lines))
fo.close()
f.close()

### SPP ###
[maxpool]
stride=1
size=5

[route]
layers=-2

[maxpool]
stride=1
size=9

[route]
layers=-4

[maxpool]
stride=1
size=13

[route]
layers=-1,-3,-5,-6

### End SPP ###