基于SIFT特征的图像特征提取

    import cv2  
    import numpy as np  
    img = cv2.imread('home.jpg')  
    gray= cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)  
    sift = cv2.SIFT()  
    kp = sift.detect(gray,None)  
    img=cv2.drawKeypoints(gray,kp)  
    cv2.imwrite('sift_keypoints.jpg',img)

    # Developer：Fazzie  
    # Time: 2020/6/1420:25  
    # File name: SIFT.py  
    # Development environment: Anaconda Python  
      
    import cv2  
    import numpy as np  
    import scipy  
    import _pickle as pickle  
    import random  
    import os  
    from matplotlib import pyplot as plt  
    import scipy.spatial  
      
    # 特征提取模块  
    def feature_extract(image_path,vector_size=32):  
        img = cv2.imread(image_path,1)#读取图片  
        # 显示图片  
        # cv2.imshow('image',img)  
        # cv2.waitKey(0)  
        # cv2.destroyAllWindows()  
      
        # SIFT特征提取  
        gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)  
        sift = cv2.xfeatures2d.SIFT_create()  
     kp = sift.detect(gray, None)  
      
      # 画出特征点并保存  
        img = cv2.drawKeypoints(gray, kp, img)  
        cv2.imwrite('sift_keypoints.jpg', img)  
      
        # 根据关键点的返回值进行排序（越大越好）  
        kp = sorted(kp, key=lambda x: -x.response)[:vector_size]  
      
        # 计算描述符向量  
        kp, des = sift.compute(gray, kp)  
      
        # 将其放在一个大的向量中，作为我们的特征向量  
        des = des.flatten()  
      
        # 使描述符的大小一致  
        # 描述符向量的大小为128  
        needed_size = (vector_size * 128)  
        if des.size < needed_size:  
            # 如果少于32个描述符，则在特征向量后面补零  
            des = np.concatenate([des, np.zeros(needed_size - des.size)])  
      
        return des  
      
    #   数据存储  
    def batch_extractor(images_path, pickled_db_path="features.pck"):  
        files = [os.path.join(images_path, p) for p in sorted(os.listdir(images_path))]  
        result = {}  
        for f in files:  
            print('Extracting features from image %s' % f)  
            name = f.split('/')[-1].lower()  
            result[name] = feature_extract(f)  
      
        # 将特征向量存于pickled 文件  
        with open(pickled_db_path, 'wb') as fp:  
            pickle.dump(result, fp)  
      
    # 图像特征匹配模块  
    class Matcher(object):  
      
        def __init__(self, pickled_db_path="features.pck"):  
            with open(pickled_db_path,"rb") as fp:  
                self.data = pickle.load(fp)  
            self.names = []  
            self.matrix = []  
            for k, v in self.data.items():  
                self.names.append(k)  
                self.matrix.append(v)  
            self.matrix = np.array(self.matrix)  
            self.names = np.array(self.names)  
      
        def cos_cdist(self, vector):  
            # 计算待搜索图像与数据库图像的余弦距离  
            v = vector.reshape(1, -1)  
            return scipy.spatial.distance.cdist(self.matrix, v, 'cosine').reshape(-1)  
      
        def match(self, image_path, topn=5):  
            features = feature_extract(image_path)  
            img_distances = self.cos_cdist(features)  
            # 获得前5个记录  
            nearest_ids = np.argsort(img_distances)[:topn].tolist()  
      
            nearest_img_paths = self.names[nearest_ids].tolist()  
            return nearest_img_paths, img_distances[nearest_ids].tolist()  
      
      
    def show_img(path):  
        img = cv2.imread(path,1)  
        plt.imshow(img)  
        plt.show()  
      
    # 主程序  
    def run():  
        images_path = 'E:/Code/SFIT/101_ObjectCategories/panda/'  
        files = [os.path.join(images_path, p) for p in sorted(os.listdir(images_path))]  
        # 随机获取1张图  
      
        sample = random.sample(files, 1)  
      
        batch_extractor(images_path)  
      
        ma = Matcher('features.pck')  
      
        for s in sample:  
            print('Query image ==========================================')  
            show_img(s)  
            names, match = ma.match(s, topn=3)  
            print('Result images ========================================')  
            for i in range(3):  
      
                # 我们得到了余弦距离，向量之间的余弦距离越小表示它们越相似，因此我们从1中减去它以得到匹配值  
                print('Match %s' % (1 - match[i]))  
                show_img(os.path.join(images_path, names[i]))  
      
    run()