主页 / 问题 / 1258108
Accepted
Анатолий Михеенко
Asked:2022-03-19 21:04:09 +0000 UTC2022-03-19 21:04:09 +0000 UTC

如何获取图像中物体的坐标？

  • 772

有一个脚本可以拍摄第一个屏幕截图

def screenshot_one(self):
        image_1 = pyautogui.screenshot(region = (5, 169, 1427, 559))
        image_1.save(r'{}\data\screenshot\image_1.png'.format(self.BASE_DIR))
        return image_1

在此处输入图像描述

第二个截图

def screenshot_two(self):
        image_2 = pyautogui.screenshot(region = (5, 169, 1427, 559))
        image_2.save(r'{}\data\screenshot\image_2.png'.format(self.BASE_DIR))
        return image_2

在此处输入图像描述

然后找出它们之间的区别

def comparison(self, image_1, image_2):
        result = ImageChops.difference(image_1, image_2)
        result.save(r'{0}\data\screenshot\screen.png'.format(self.BASE_DIR))

在此处输入图像描述

如何获取1920х1080脚本突出显示的对象在屏幕上的坐标作为第一张和第二张截图之间的差异，即第三张截图上白色圆圈的坐标？

整个代码

# -*- coding: utf-8 -*-

from PIL import Image, ImageChops
from pathlib import Path
import pyautogui
import keyboard
import time

class Object_search:
    def __init__(self):
        self.BASE_DIR = Path(__file__).resolve().parent

    def screenshot_one(self):
        image_1 = pyautogui.screenshot(region = (5, 169, 1427, 559))
        image_1.save(r'{}\data\screenshot\image_1.png'.format(self.BASE_DIR))
        return image_1

    def screenshot_two(self):
        image_2 = pyautogui.screenshot(region = (5, 169, 1427, 559))
        image_2.save(r'{}\data\screenshot\image_2.png'.format(self.BASE_DIR))
        return image_2

    def comparison(self, image_1, image_2):
        result = ImageChops.difference(image_1, image_2)
        result.save(r'{0}\data\screenshot\screen.png'.format(self.BASE_DIR))

while True:
    if keyboard.is_pressed('5'):
        print('image_1')
        image_1 = Object_search().screenshot_one()
        time.sleep(0.1)
    elif keyboard.is_pressed('6'):
        image_2 = Object_search().screenshot_two()
        print('image_2')
        time.sleep(0.1)
    elif keyboard.is_pressed('7'):
        print('comparison')
        comparison = Object_search().comparison(image_1, image_2)
        time.sleep(0.1)
    elif keyboard.is_pressed('8'):
        print('exit')
        break
python

1 个回答

  1. Best Answer
    img = cv2.imread("MYIMG.png", cv2.IMREAD_GRAYSCALE)
(_, _, _, coords) = cv2.minMaxLoc(img)

    coords - 圆的左上角。

    UPD

    我从旧项目中找到了另一种选择：

    import cv2

img = cv2.imread("AFQZB.png", cv2.IMREAD_GRAYSCALE)
# img=cv2.cvtColor(self.image, cv2.COLOR_BGR2GRAY)

(_, threshold)=cv2.threshold(img, 45, 255, cv2.THRESH_BINARY)
(contours, _)=cv2.findContours(threshold, cv2.RETR_CCOMP, cv2.CHAIN_APPROX_SIMPLE)

for rect in contours:
    (x, y, width, height)=cv2.boundingRect(rect)
    print(x, y, width, height)
    # ...

    结论：

    411 158 222 214
    • 1

相关问题