# 提取特征并保存 def get_feat(image_list, name_list, label_list, savePath): i = 0 for image in image_list: try: # 如果是灰度图片 把3改为-1 image = np.reshape(image, (image_height, image_width, 3)) except: print('发送了异常，图片大小size不满足要求：',name_list[i]) continue gray = rgb2gray(image) / 255.0 # 这句话根据你的尺寸改改 fd = hog(gray, orientations=12,block_norm='L1', pixels_per_cell=[8, 8], cells_per_block=[4, 4], visualize=False, transform_sqrt=True) fd = np.concatenate((fd, [label_list[i]])) fd_name = name_list[i] + '.feat' fd_path = os.path.join(savePath, fd_name) joblib.dump(fd, fd_path) i += 1 print("Test features are extracted and saved.")加入颜色特征代码

以下是添加颜色特征的代码示例：

import cv2
import glob
import os
import numpy as np
from skimage.feature import hog
from skimage.color import rgb2gray
from sklearn.externals import joblib

def get_feat(image_list, name_list, label_list, savePath):
    i = 0
    for image in image_list:
        try:
            # 如果是灰度图片  把3改为-1
            image = np.reshape(image, (image_height, image_width, 3))
        except:
            print('发送了异常，图片大小size不满足要求：',name_list[i])
            continue
        gray = rgb2gray(image) / 255.0
        # 添加颜色特征
        hsv = cv2.cvtColor(image, cv2.COLOR_RGB2HSV)
        color_hist = cv2.calcHist([hsv], [0, 1, 2], None, [8, 8, 8], [0, 180, 0, 256, 0, 256])
        color_hist = cv2.normalize(color_hist, color_hist).flatten()
        # 使用 HOG 特征
        hog_feat = hog(gray, orientations=12, block_norm='L1', pixels_per_cell=[8, 8], cells_per_block=[4, 4], visualize=False, transform_sqrt=True)
        # 组合特征向量
        feat = np.concatenate((hog_feat, color_hist, [label_list[i]]))
        feat_name = name_list[i] + '.feat'
        feat_path = os.path.join(savePath, feat_name)
        joblib.dump(feat, feat_path)
        i += 1
    print("Features are extracted and saved.")

在这个示例中，首先使用 OpenCV 的 cvtColor 函数将图像从 RGB 转换为 HSV，然后使用 calcHist 函数计算 HSV 颜色直方图。接下来，使用 normalize 函数对颜色直方图进行归一化，并将其展平为一维数组。最后，将 HOG 特征和颜色特征连接起来，形成最终的特征向量。