# encoding: utf-8
import cv2 as cv
import numpy as np
import random

from ssat_sdk.picture.RGB import RGBColor

'''
用于图片初级检测，例如：颜色、噪声
此类已经废弃
'''
def getMaxRGB(imgFile,area):
    img = cv.imread(imgFile,1)
    if (area is None):
        area = (0,0,img.shape[1], img.shape[0])#mat shape:高、宽、颜色值个数
    img = cv.cvtColor(img,cv.COLOR_BGR2RGB)
    img = np.array(img)
    if len(area) == 4:
        rgbColor = RGBColor()
        x1 = area[0]
        y1 = area[1]
        x2 = area[2]
        y2 = area[3]
        img = img[y1:y2+1,x1:x2+1]
        return rgbColor.getMaxBGR(img)

    else:
        return None,None,None

'''
截取图片区域RGB平均值。
:param area:二维区域。坐标系：x轴：水平向右，y轴：纵向向下
'''
def getAverageRGB(img_addr, area = None):
    img = cv.imread(img_addr)
    return getImgAverageRGB(img, area)

'''
截取图片区域RGB平均值。
:param area:二维区域。坐标系：x轴：水平向右，y轴：纵向向下
'''
def getImgAverageRGB(img, area):
    if (area is None):
        area = (0,0,img.shape[1], img.shape[0])#mat shape:高、宽、颜色值个数
    print "getImgAverageRGB", img.shape, area
    # print img.shape
    img = cv.cvtColor(img,cv.COLOR_BGR2RGB)
    img = np.array(img)
    if len(area) == 4:
        rgbColor = RGBColor()
        x1 = area[0]
        y1 = area[1]
        x2 = area[2]
        y2 = area[3]
        img = img[y1:y2, x1:x2]
        return rgbColor.getAvgBGR(img)

    else:
        return None, None, None

'''@废弃'''
# 判断黑屏
def isBlack(img_addr, black_th=15):
    gray = cv.imread(img_addr,0)
    width = gray.shape[1]
    hight = gray.shape[0]

    for i in range(0,50):
        random_y = random.randint(0,width-1)
        random_x = random.randint(0,hight-1)
        if gray[random_x][random_y] >= black_th:
            return False
        else:
            continue
    return True


 #判断噪点
def hasNoise(img_addr):
    pass



def getPoints(img_addr):
    kernel = np.ones((11, 11), dtype=np.uint8)
    img = cv.imread(img_addr, 1)
    gray = cv.cvtColor(img, cv.COLOR_BGR2GRAY)
    gray = gray[gray.shape[0]/2:gray.shape[0],gray.shape[1]/2:gray.shape[1]]
    x = cv.Sobel(gray, cv.CV_16S,1, 0)
    y = cv.Sobel(gray, cv.CV_16S, 0, 1)
    absX = cv.convertScaleAbs(x)   # 转回uint8
    absY = cv.convertScaleAbs(y)
    dst = cv.addWeighted(absX, 0.5, absY, 0.5,0)
    _, dst_2 = cv.threshold(dst, 100, 255, cv.THRESH_BINARY)
    img_close = cv.erode(cv.dilate(dst_2, kernel), kernel)
    image, contours, hierarchy = cv. findContours(img_close, 1, 2)
    true_rectang = []
    for i in range(0,len(contours)):
        rectang = []
        temp_arr = []
        for j in range(0,len(contours[i])):
            for points_index in range(0, len(contours[i][j])):
                temp_arr.append(contours[i][j][0])
        for temp_arr_index in range(0, len(temp_arr)):
            if temp_arr_index == 0:
                # 第一个点加入数组
                rectang.append(temp_arr[temp_arr_index])
            else:
                # 遍历所有点
                flag = 0
                for rectang_index in range(0, len(rectang)):
                    if (abs(temp_arr[temp_arr_index][0]-rectang[rectang_index][0]) >= 15) or (abs(temp_arr[temp_arr_index][1]-rectang[rectang_index][1]) >= 15):
                       flag = flag + 1
                    if flag == len(rectang):
                        rectang.append(temp_arr[temp_arr_index])
        if len(rectang) == 4:
            # 判断是否我们需要的矩形 ，如果是 则返回到true_rectang
            true_rectang.append(rectang)
    return true_rectang


'''
@废弃
判断马赛克
'''
def hasMosaic(std_pic,test_pic,num):


    std_pic = getPoints(std_pic)
    test_pic = getPoints(test_pic)

    print len(std_pic)
    print len(test_pic)

    if len(test_pic) != len(std_pic):
        return True
    else:
        return False

class PreImageJudge():
    def __init__(self):
        self.BGR = RGBColor()

    '''
    判断一张图片是否是纯黑图片。
    :param img:需要判断的图片
    :param grayTH : 黑色判断的灰阶阈值
    :param splitCount: 图片拆分的行列次数
    :param expTh：黑色计算的容错区域比例。
    '''
    def isBlack(self, img, grayTH = 70,splitCount=10, expTh = 0.01):
        return self.BGR.isBlack()

if __name__ == '__main__':
    print hasMosaic('std.jpg','test.jpg',num=None)