def search(img_path): cf = color.get_color_feature(img_path) cfs = File_Operation.read_list_from_file('/Users/ligang/Documents/cfs.txt') distances = [] for cf_tuple in cfs: d = Distance.distance(cf, cf_tuple[1]) distances.append((cf_tuple[0], d)) top_distances = heapq.nsmallest(10, distances, key=lambda x: x[1]) print top_distances
def search(img_path): cf = color.get_color_feature(img_path) unlbp_feature = ulbp.ulbp(img_path) cfs = File_Operation.read_list_from_file('/Users/ligang/Documents/cfs.txt') ulbps = File_Operation.read_list_from_file('/Users/ligang/Documents/ulbp.txt') distances = [] for cf_tuple, ulbp_tuple in zip(cfs, ulbps): assert cf_tuple[0] == ulbp_tuple[0] d_color = Distance.distance(cf, cf_tuple[1]) d_ulbp = Distance.distance(unlbp_feature, ulbp_tuple[1]) d = d_color + d_ulbp distances.append((cf_tuple[0], d)) top_distances = heapq.nsmallest(20, distances, key=lambda x: x[1]) dstDir = '/Users/ligang/Documents/Emilie/colorlbp_search_result' img_set_dir = '/Users/ligang/Documents/Emilie/dress' for top in top_distances: shutil.copy2(os.path.join(img_set_dir, top[0]), os.path.join(dstDir, top[0])) print top_distances