如何在屏幕上查找具有透明背景的图像？

如何解决如何在屏幕上查找具有透明背景的图像？

因此，我有一幅具有透明背景和尺寸（Width_1，High_1）的图像，而我想做的就是找到该图像（但不必使用相同的尺寸）在屏幕上以尺寸（Width_2，High_2）来表示，因此我谨记以下代码：

def resize_untill_find(image,bigger_or_smaller,min_size = 50,max_size = 150,one_match_per_point = 'on'): #(imagem,diminuir ou aumentar,min_size = reduce the image until min_size%,max_size = expand the image until max_size%)
    print('Press CTRL+ALT to start:')
    while True:
        if keyboard.is_pressed('control+alt'):
            break
    image_info = Ferramentas.get_image_first_info(image)# get width,high,and pixels number off the image
    image = cv2.imread(image)
    image_rate = image_info['whidth'] / image_info['high'] #proportion Width/High
    ocurrencies = {} # [x,y,w,h]
    match_number = 1 # number of the ocurrency
    high = image_info['high']
    if bigger_or_smaller == 'smaller':
        while int((high/image_info['high'])*100) >= int(min_size): # while the resize are not under min_size
            print('{}%'.format(round((high/image_info['high'])*100)))
            width = round(image_info['whidth']-((image_info['high'] - high) * image_rate))
            dimension = (width,high) # new dimension of the image
            print('Dimension:',dimension)
            image_resized = cv2.resize(image,dimension) #resize the image
                matches_position = pyautogui.locateallOnScreen(image_resized,confidence = 0.9)
                for region in matches_position:
                    region = list(region)
                    ocurrencies['Match_{}'.format(match_number)] = region # region = [X,Y,new_Width,new_High]
                    match_number += 1 
            if one_match_per_point == 'on':
                match_position = pyautogui.locateOnScreen(image_resized,confidence = 0.9)
                if match_position != None:
                    match_position = list(match_position)
                    ocurrencies['Match_{}'.format(match_number)] = match_position # region = [X,new_High]
                    match_number += 1
            high -= 1
    if bigger_or_smaller == 'bigger':
        while int((high/image_info['high'])*100) <= int(max_size):
            print('{}%'.format(round((high/image_info['high'])*100)))
            width = round(image_info['whidth']+((high - image_info['high']) * image_rate))
            dimension = (width,high) # image new dimension
            print('Dimension:',dimension) #resize the imge with new dimensions
            if one_match_per_point == 'off':
                matches_position = pyautogui.locateallOnScreen(image_resized,confidence = 0.9) # lista todos os matches 90% iguais ao da imagem redimensionada
                for region in matches_position:
                    region = list(region)
                    ocurrencies['Match_{}'.format(match_number)] = region # region = [X,new_High]
                    match_number += 1   
            if one_match_per_point == 'on':
                match_position = pyautogui.locateOnScreen(image_resized,new_High]
                    match_number += 1 
            high += 1
    print(ocurrencies)
    cv2.waitKey(0)
    return ocurrencies

我不知道是否有更好的方法来执行此操作，如果有，请告诉我，我将很高兴改进代码（它必须尽可能快），但是主要问题是：它无法匹配具有透明度和透明度渐变背景的模板。

所以我有这张图片：

，程序无法找到图像并获取坐标。

* 编辑：

所以我可以拍摄这张照片：

并调整大小，直到在此图像上找到匹配项：

但是当我尝试用透明背景拍摄.png图片时，它不起作用：

（模板），然后尝试在此模板上找到匹配项

（ screen1 ）或此

（ screen2 ）。

您会注意到，第一个模板的尺寸与 screen1 相同，但是背景为深灰色，而 screen2 的背景也不同，并且较小（但保持原始的“宽/高”比例）。

所以我想要的是能够使用 template 在屏幕的某个位置找到 screen1 或 screen2 。 / p>

解决方法

def search(screen,img):
sx,sy = screen.size
ix,iy = img.size
for xstart in range(sx - ix): 
    for ystart in range(sy - iy):
        #search for the pixel on the screen that equals the pixel at img[0:0]
        if img.getpixel((0,0)) == screen.getpixel((xstart,ystart)):
            match = 1 #temporary
            for x in range(ix): #check if first row of img is on this coords
                if img.getpixel((x,0)) <> screen.getpixel((xstart+x,ystart)):
                    match = 0 #if there's any difference,exit the loop
                    break 
            if match == 1: #otherwise,if this coords matches the first row of img
                for x in range(ix): 
                    for y in range(iy):
                        #check every pixel of the img
                        if img.getpixel((x,y)) <> screen.getpixel((xstart+x,ystart+y)):
                            match = 0 #any difference - break
                            break
                if match == 1: return (xstart,ystart) #return top-left corner coordinates
return (-1,-1) #or this,if not found
#i found this somewhere else