6,研发项目 » KI007 智能终端（刘俊锋）

import argparse

import os

import cv2

import numpy as np

import time

def ImageRotate(img, angle):   # img:输入图片；newIm：输出图片；angle：旋转角度(°)

    height, width = img.shape[:2]  # 输入(H,W,C)，取 H，W 的值

    center = (width // 2, height // 2)  # 绕图片中心进行旋转

    M = cv2.getRotationMatrix2D(center, angle, 1.0)

    image_rotation = cv2.warpAffine(img, M, (width, height))

    return image_rotation

def circle_tr(src):

    dst = np.zeros(src.shape, np.uint8)  # 感兴趣区域ROI

    mask = np.zeros(src.shape, dtype='uint8')  # 感兴趣区域ROI

    (h, w) = mask.shape[:2]

    (cX, cY) = (w // 2, h // 2)  # 是向下取整

    radius = int(min(h, w) / 2)

    cv2.circle(mask, (cX, cY), radius, (255, 255, 255), -1)

    # 以下是copyTo的算法原理：

    # 先遍历每行每列（如果不是灰度图还需遍历通道，可以事先把mask图转为灰度图）

    for row in range(mask.shape[0]):

        for col in range(mask.shape[1]):

            # 如果掩图的像素不等于0，则dst(x,y) = scr(x,y)

            if mask[row, col] != 0:

                # dst_image和scr_Image一定要高宽通道数都相同，否则会报错

                dst[row, col] = src[row, col]

                # 如果掩图的像素等于0，则dst(x,y) = 0

            elif mask[row, col] == 0:

                dst[row, col] = 0

    return dst

def ImagePyrDown(image,NumLevels):

    for i in range(NumLevels):

        image = cv2.pyrDown(image)       #pyrDown下采样

    return image

# 旋转匹配函数（输入参数分别为模板图像、待匹配图像）

def RatationMatch(modelpicture, searchpicture, Layer):

    searchtmp = []

    modeltmp = []

    searchtmp = ImagePyrDown(searchpicture, Layer)

    modeltmp = ImagePyrDown(modelpicture, Layer)

    newIm = circle_tr(modeltmp)

    # 使用matchTemplate对原始灰度图像和图像模板进行匹配

    res = cv2.matchTemplate(searchtmp, newIm, cv2.TM_SQDIFF_NORMED)

    min_val, max_val, min_indx, max_indx = cv2.minMaxLoc(res)

    location = min_indx

    temp = min_val

    angle = 0  # 当前旋转角度记录为0

    # 以步长为5进行第一次粗循环匹配

    for i in range(-180, 181, 5):

        newIm= ImageRotate(modeltmp, i)

        newIm = circle_tr(newIm)

        res = cv2.matchTemplate(searchtmp, newIm, cv2.TM_SQDIFF_NORMED)

        min_val, max_val, min_indx, max_indx = cv2.minMaxLoc(res)

        if min_val < temp:

            location = min_indx

            temp = min_val

            angle = i

    # 在当前最优匹配角度周围10的区间以1为步长循环进行循环匹配计算

    for j in range(angle - 5, angle + 6):

        newIm = ImageRotate(modeltmp, j)

        newIm = circle_tr(newIm)

        res = cv2.matchTemplate(searchtmp, newIm, cv2.TM_SQDIFF_NORMED)

        min_val, max_val, min_indx, max_indx = cv2.minMaxLoc(res)

        if min_val < temp:

            location = min_indx

            temp = min_val

            angle = j

    # 在当前最优匹配角度周围2的区间以0.1为步长进行循环匹配计算

    k_angle = angle - 0.9

    for k in range(0, 19):

        k_angle = k_angle + 0.1

        newIm = ImageRotate(modeltmp, k_angle)

        newIm = circle_tr(newIm)

        res = cv2.matchTemplate(searchtmp, newIm, cv2.TM_SQDIFF_NORMED)

        min_val, max_val, min_indx, max_indx = cv2.minMaxLoc(res)

        if min_val < temp:

            location = min_indx

            temp = min_val

            angle = k_angle

    # 用下采样前的图片来进行精匹配计算

    k_angle = angle - 0.1

    newIm = ImageRotate(modelpicture, k_angle)

    newIm = circle_tr(newIm)

    res = cv2.matchTemplate(searchpicture, newIm, cv2.TM_CCOEFF_NORMED)

    min_val, max_val, min_indx, max_indx = cv2.minMaxLoc(res)

    location = max_indx

    temp = max_val

    angle = k_angle

    for k in range(1, 3):

        k_angle = k_angle + 0.1

        newIm= ImageRotate(modelpicture, k_angle)

        newIm = circle_tr(newIm)

        res = cv2.matchTemplate(searchpicture, newIm, cv2.TM_CCOEFF_NORMED)

        min_val, max_val, min_indx, max_indx = cv2.minMaxLoc(res)

        if max_val > temp:

            location = max_indx

            temp = max_val

            angle = k_angle

    location_x = location[0] #+ 50

    location_y = location[1] #+ 50

    # 前面得到的旋转角度是匹配时模板图像旋转的角度，后面需要的角度值是待检测图像应该旋转的角度值，故需要做相反数变换

    angle = -angle

    match_point = {'angle': angle, 'point': (location_x, location_y)}

    return match_point

# 画图

def draw_result(src, temp, match_pointp, match_pointa):

    height, width = temp.shape[:2]  # 输入(H,W,C)，取 H，W 的值

    size = (height,width)

    center = (match_pointp[0]+width/2, match_pointp[1]+height/2)  # 绕图片中心进行旋转

    M = cv2.getRotationMatrix2D(center, -match_pointa, 1.0)

    rect_points = cv2.boxPoints(((center[0], center[1]), size, 0))

    rect_points = np.int0(cv2.transform(np.array([rect_points]), M))[0]

    cv2.polylines(src, [rect_points], isClosed=True, color=(0, 255, 0), thickness=2)

    #cv2.rectangle(src, match_pointp,

                  #(match_pointp[0] + temp.shape[1], match_pointp[1] + temp.shape[0]),

                  #(0, 255, 0), 2)

    #cv2.imshow('result', src)

    #cv2.waitKey()

    print("坐标",center)

def oimage_read_from_chinese_path(ppath, TemplatePath, Layer):

    img = cv2.imdecode(np.fromfile(ppath,

                                   dtype=np.uint8), 1)

    templeimg = cv2.imdecode(np.fromfile(TemplatePath,

                                         dtype=np.uint8), 1)

    ModelImage = templeimg

    SearchImage = img

    ModelImage_edge = cv2.GaussianBlur(ModelImage, (5, 5), 0)

    ModelImage_edge = cv2.Canny(ModelImage_edge, 10, 200, apertureSize=3)

    SearchImage_edge = cv2.GaussianBlur(SearchImage, (5, 5), 0)

    (h1, w1) = SearchImage_edge.shape[:2]

    SearchImage_edge = cv2.Canny(SearchImage_edge, 10, 180, apertureSize=3)

    serch_ROIPart = SearchImage_edge[50:h1 - 50, 50:w1 - 50]  # 裁剪图像

    match_points = RatationMatch(ModelImage_edge, SearchImage_edge, Layer)

    #TmpImage_edge = ImageRotate(SearchImage_edge, match_points['angle'])

    #cv2.imshow("TmpImage_edge", TmpImage_edge)

    #cv2.waitKey()

    draw_result(SearchImage, ModelImage_edge, match_points['point'], match_points['angle'])

    print("角度", -match_points['angle'])

    return SearchImage

def print_hi(imgpath, Process, Xstart, Ystart, Xend, Yend, Layer):

    if Process == 1:

        TemplatePath = "croppedimage.bmp"

        img_path = oimage_read_from_chinese_path(imgpath, TemplatePath, Layer)

        cv2.imwrite("OutImage.bmp", img_path)

    elif Process == 0:

        image_numpy_data = cv2.imdecode(np.fromfile(imgpath, dtype=np.uint8), 1)

        #image_numpy_data = cv2.imread(imgpath)

        cropped_image = image_numpy_data[Ystart:Yend, Xstart:Xend]

        cv2.rectangle(image_numpy_data, (Xstart,Ystart),

                      (Xend, Yend),

                      (0, 255, 0), 2)

        cv2.imwrite("croppedimage.bmp", cropped_image)

        cv2.imwrite("OutImage.bmp", image_numpy_data)

# Press the green button in the gutter to run the script.

if __name__ == '__main__':

    parser = argparse.ArgumentParser()

    parser.add_argument("--ImagePath", type=str, required=True)

    parser.add_argument("--Process", type=int, required=True)

    parser.add_argument("--Xstart", type=int, required=True)

    parser.add_argument("--Ystart", type=int, required=True)

    parser.add_argument("--Xend", type=int, required=True)

    parser.add_argument("--Yend", type=int, required=True)

    parser.add_argument("--Layer", type=int, required=True)

    args = parser.parse_args()

    print_hi(args.ImagePath, args.Process, args.Xstart, args.Ystart, args.Xend, args.Yend,  args.Layer)