python 测试速度（cv2 vs scipy vs tensorflow）

Posted 2021-05-09

# Test which library read and resize JPG files faster
# cv2 vs scipy

import numpy as np
np.random.seed(2016)

import os
import glob
import cv2
import math
import time
from scipy.misc import imread, imresize
import tensorflow as tf


def get_im_skipy(path, img_rows, img_cols, color_type=1):
    # Load as grayscale
    if color_type == 1:
        img = imread(path, True)
    elif color_type == 3:
        img = imread(path)
    # Reduce size
    resized = imresize(img, (img_cols, img_rows))
    return resized


def get_im_cv2(path, img_rows, img_cols, color_type=1):
    # Load as grayscale
    if color_type == 1:
        img = cv2.imread(path, 0)
    elif color_type == 3:
        img = cv2.imread(path)
    # Reduce size
    resized = cv2.resize(img, (img_cols, img_rows))
    return resized


def get_im_tf(path, img_rows, img_cols, color_type=1):
    img = tf.read_file(path)
    # Load as grayscale
    if color_type == 1:
        img = tf.image.decode_jpeg(img, channels=1)
    elif color_type == 3:
        img = tf.image.decode_jpeg(img, channels=0)
    # Reduce size
    resized = tf.image.resize_images(img, (img_rows, img_cols))
    return resized


def get_driver_data():
    dr = dict()
    path = os.path.join('..', 'input', 'driver_imgs_list.csv')
    print('Read drivers data')
    f = open(path, 'r')
    line = f.readline()
    while (1):
        line = f.readline()
        if line == '':
            break
        arr = line.strip().split(',')
        dr[arr[2]] = arr[0]
    f.close()
    return dr


def load_train(img_rows, img_cols, color_type=1, type=0):
    X_train = []
    y_train = []
    driver_id = []

    driver_data = get_driver_data()

    print('Read train images')
    for j in range(10):
        print('Load folder c{}'.format(j))
        path = os.path.join('..', 'input', 'train', 'c' + str(j), '*.jpg')
        files = glob.glob(path)
        for fl in files:
            flbase = os.path.basename(fl)
            if type == 0:
                img = get_im_cv2(fl, img_rows, img_cols, color_type)
            elif type == 1:
                img = get_im_skipy(fl, img_rows, img_cols, color_type)
            else:
                img = get_im_tf(fl, img_rows, img_cols, color_type)
            X_train.append(img)
            y_train.append(j)
            driver_id.append(driver_data[flbase])

    unique_drivers = sorted(list(set(driver_id)))
    print('Unique drivers: {}'.format(len(unique_drivers)))
    print(unique_drivers)
    return X_train, y_train, driver_id, unique_drivers


def load_test(img_rows, img_cols, color_type=1, type=0):
    print('Read test images')
    path = os.path.join('..', 'input', 'test', '*.jpg')
    files = glob.glob(path)
    files = files[:len(files) // 2]
    X_test = []
    X_test_id = []
    total = 0
    thr = math.floor(len(files)/10)
    for fl in files:
        flbase = os.path.basename(fl)
        if type == 0:
            img = get_im_cv2(fl, img_rows, img_cols, color_type)
        elif type == 1:
            img = get_im_skipy(fl, img_rows, img_cols, color_type)
        else:
            img = get_im_tf(fl, img_rows, img_cols, color_type)
        X_test.append(img)
        X_test_id.append(flbase)
        total += 1
        if total%thr == 0:
            print('Read {} images from {}'.format(total, len(files)))

    return X_test, X_test_id

start_time = time.time()
load_train(24, 32, 1, 2)
load_test(24, 32, 1, 2)
tf_time = time.time() - start_time
print('Tensorflow:', tf_time, ' seconds')

start_time = time.time()
load_train(24, 32, 1, 0)
load_test(24, 32, 1, 0)
cv2_time = time.time() - start_time
print('CV2:', cv2_time, ' seconds')

start_time = time.time()
load_train(24, 32, 1, 1)
load_test(24, 32, 1, 1)
sk_time = time.time() - start_time
print('Scipy:', sk_time, ' seconds')