python类NORM_L2的实例源码

def extractFeatures(self):
        if len(self.image) == 0:
            print 'Warning: No image detected. Features not extracted.'
            return None
        else:
            self.net.blobs['data'].reshape(1, 3, self.crop, self.crop)
            self.net.blobs['data'].data[...] = self.transformer.preprocess('data', self.image)
            self.net.forward()
            features = self.net.blobs[self.layer].data.copy()


            features = np.reshape(features, (features.shape[0], -1))[0]

            if cv2.norm(features, cv2.NORM_L2) > 0:
                features = features / cv2.norm(features, cv2.NORM_L2)
            return features.tolist()

def normalize_result(webcam, idcard):
    diff_correy = cv2.norm(settings.COREY_MATRIX, idcard, cv2.NORM_L2)
    diff_wilde = cv2.norm(settings.WILDE_MATRIX, idcard, cv2.NORM_L2)
    diff_min = diff_correy if diff_correy < diff_wilde else diff_wilde
    diff = cv2.norm(webcam, idcard, cv2.NORM_L2)
    score = float(diff) / float(diff_min)
    percentage = (1.28 - score * score * score) * 10000 / 128
    return {
        'percentage': percentage,
        'score': score,
        'message': utils.matching_message(score)
    }

def create_vse(vocabulary_path, recognized_visual_words=1000):
    """Create visual search engine with default configuration."""
    ranker = SimpleRanker(hist_comparator=Intersection())
    inverted_index = InvertedIndex(ranker=ranker, recognized_visual_words=recognized_visual_words)
    bag_of_visual_words = BagOfVisualWords(extractor=cv2.xfeatures2d.SURF_create(),
                                           matcher=cv2.BFMatcher(normType=cv2.NORM_L2),
                                           vocabulary=load(vocabulary_path))
    return VisualSearchEngine(inverted_index, bag_of_visual_words)

def build_filters(self, w, h,num_theta, fi, sigma_x, sigma_y, psi):
        "Get set of filters for GABOR"
        filters = []
        for i in range(num_theta):
            theta = ((i+1)*1.0 / num_theta) * np.pi
            for f_var in fi:
                kernel = self.get_gabor_kernel(w, h,sigma_x, sigma_y, theta, f_var, psi)
                kernel = 2.0*kernel/kernel.sum()
                # kernel = cv2.normalize(kernel, kernel, 1.0, 0, cv2.NORM_L2)
                filters.append(kernel)
        return filters

def distanceOfFV(self, fv1, fv2):
        "distance of feature vector 1 and feature vector 2"
        normset = []
        for i in range(len(fv1)):
            k = fv1[i]
            p = fv2[i]
            # k = cv2.normalize(fv1[i],k,1.0,0,norm_type=cv2.NORM_L2)
            # p = cv2.normalize(fv2[i],p,1.0,0,norm_type=cv2.NORM_L2)
            normset.append((p-k)**2.0)
        sums = 0
        sums = sum([i.sum() for i in normset])
        return mth.sqrt(sums)/100000