# coding=utf8
from PIL import ImageFilter, Image, ImageStat, ImageEnhance
import PIL
import numpy
import numpy as np
import os
import math
import cv2

filepath = '/Users/Ikaros/Downloads/20160729_晚上'
image_enhance_parameter = {
    "brightness": 2,
    "contrast": 2
}


def brightness(image):
    stat = ImageStat.Stat(image)
    if len(stat.mean) == 3:
        r, g, b = stat.mean
        return math.sqrt(0.241 * (r**2) + 0.691 * (g**2) + 0.068 * (b**2))
    else:
        return stat.mean[0]


def histeq(im,nbr_bins=256):

    # get image histogram
    imhist,bins = numpy.histogram(numpy.array(im).reshape(-1),nbr_bins,normed=True)
    cdf = imhist.cumsum() #cumulative distribution function
    cdf = 255 * cdf / cdf[-1] #normalize

    # use linear interpolation of cdf to find new pixel values
    im2 = numpy.interp(numpy.array(im).reshape(-1),bins[:-1],cdf)

    return im2.reshape(numpy.array(im).shape), cdf


def denoise_image(image):
    return image.filter(ImageFilter.MedianFilter(3))


def enhance_image(image):
    # image = image.convert("L")
    contrast_converter = ImageEnhance.Contrast(image)
    image = contrast_converter.enhance(image_enhance_parameter['contrast'])

    brightness_converter = ImageEnhance.Brightness(image)
    image = brightness_converter.enhance(image_enhance_parameter['brightness'])

    return image


def binarize_image(numpy_array_like_image, threshold=65,
                   output_value_above_threshold=None, output_value_below_threshold=None):
    if output_value_above_threshold is None:
        output_value_above_threshold = numpy_array_like_image
    if output_value_below_threshold is None:
        output_value_below_threshold = numpy_array_like_image
    return numpy.where(numpy_array_like_image > threshold,
                       output_value_above_threshold, output_value_below_threshold)


def adjust_gamma(image, gamma=1.0):
    # build a lookup table mapping the pixel values [0, 255] to
    # their adjusted gamma values
    invGamma = 1.0 / gamma
    table = np.array([((i / 255.0) ** invGamma) * 255
                      for i in np.arange(0, 256)]).astype("uint8")

    # apply gamma correction using the lookup table
    return cv2.LUT(image, table)


def main():
    filelist = [os.path.join(filepath, file) for file in os.listdir(filepath)]
    file_count = 1
    for f in filelist:
        if '.jpg' not in f and '.JPG' not in f:
            continue
        img = Image.open(f)

        # car_region for 720x576
        # car_region = (50,240,450,400)

        # car_region for 1024x768
        car_region = (50, 300, 640, 580)
        img = img.crop(car_region)
        # print "The brightness value of", f, "is", brightness(img)


        img = img.convert("L")

        cv_image = numpy.asarray(img)
        # cv_image = cv2.fastNlMeansDenoising(cv_image, None, 5)
        cv2.denoise_TVL1(cv_image, cv_image)

        img = enhance_image(Image.fromarray(cv_image))

        cv_image = adjust_gamma(numpy.asarray(img), 1.5)
        cv2.imwrite(os.path.join(filepath, "%02d" % file_count + ".jpg"), cv_image)

        print f, '-->', os.path.join(filepath, "%02d" % file_count + ".jpg")
        file_count += 1


if __name__ == '__main__':
    main()