Kivy/KivyMD中如何在.kv与非.kv定义的屏幕间切换

你当前代码的根本问题是将CameraScreen定义为MDApp子类。Kivy/KivyMD 程序全局仅允许一个运行中的App实例，切换屏幕时调用CameraScreen().run()本质是启动第二个独立程序，完全不符合ScreenManager的屏幕切换逻辑。

正确实现方式是将CameraScreen改为和HomeScreen、ManageDataScreen一致的Screen子类，把摄像头相关逻辑迁移到该Screen类中，界面部分通过kv文件定义，所有屏幕统一交给ScreenManager管理。

修正步骤 #

1. 替换Python主文件代码 #

from kivymd.app import MDApp
from kivy.uix.screenmanager import Screen
from kivy.uix.image import Image
from kivy.lang import Builder
from kivy.graphics.texture import Texture
from kivy.clock import Clock
import cv2
import pytesseract


class HomeScreen(Screen):
    pass

class ManageDataScreen(Screen):
    pass

# CameraScreen改为继承Screen，不再作为独立App
class CameraScreen(Screen):
    def on_enter(self, *args):
        # 进入屏幕时初始化摄像头、启动帧读取调度
        self.capture = cv2.VideoCapture(0)
        self.video_clock = Clock.schedule_interval(self.load_video, 1.0/30.0)

    def on_leave(self, *args):
        # 离开屏幕时释放资源，避免后台占用摄像头
        self.video_clock.cancel()
        self.capture.release()

    def load_video(self, *args):
        ret, frame = self.capture.read()
        if not ret:
            return
        self.image_frame = frame
        buffer = cv2.flip(frame, 0).tobytes()
        texture = Texture.create(size=(frame.shape[1], frame.shape[0]), colorfmt='bgr')
        texture.blit_buffer(buffer, colorfmt='bgr', bufferfmt='ubyte')
        self.ids.camera_image.texture = texture

    def take_picture(self, *args):
        image_name = "img_TIMESTAMP.png"
        img = cv2.cvtColor(self.image_frame, cv2.COLOR_BGR2GRAY)
        img = cv2.GaussianBlur(img, (3, 3), 0)
        img = cv2.threshold(img, 0, 255, cv2.THRESH_BINARY_INV + cv2.THRESH_OTSU)[1]
        text_data = pytesseract.image_to_string(img, lang='eng', config="--psm 6")
        self.ids.ocr_result_label.text = text_data
        cv2.imshow("cv2 final image", img)
        cv2.imwrite(image_name, self.image_frame)

GUI = Builder.load_file("kv/main.kv")
class MainApp(MDApp):
    def build(self):
        return GUI

    def change_screen(self, screen_name):
        screen_manager = self.root.ids['screen_manager']
        # 所有屏幕统一通过ScreenManager切换，不再启动新App
        screen_manager.current = screen_name

if __name__ == '__main__':
    pytesseract.pytesseract.tesseract_cmd = r"C:\Program Files\Tesseract-OCR\tesseract.exe"
    MainApp().run()

2. 修改kv/main.kv文件 #

添加摄像头屏幕的kv引入，同时将CameraScreen注册到ScreenManager中：

#:include kv/home_screen.kv
#:include kv/manage_data_screen.kv
#:include kv/camera_screen.kv

GridLayout:
    cols: 1
    ScreenManager:
        id: screen_manager
        HomeScreen:
            name: "home_screen"
            id: home_screen
        ManageDataScreen:
            name: "manage_data_screen"
            id: manage_data_screen
        CameraScreen:
            name: "camera_screen"
            id: camera_screen

3. 新建kv/camera_screen.kv文件 #

将原Python代码中build方法内的界面布局转为kv语法，控件id和Python代码中的引用保持一致：

<CameraScreen>:
    MDBoxLayout:
        orientation: "vertical"
        Image:
            id: camera_image
        MDLabel:
            id: ocr_result_label
            text: ""
        MDRaisedButton:
            text: "Capture"
            pos_hint: {"center_x": .5, "center_y": .5}
            size_hint: None, None
            on_press: root.take_picture()

注意事项 #

• 所有屏幕切换直接调用app.change_screen("目标屏幕name")即可，比如切摄像头屏传"camera_screen"，返回主界面传"home_screen"
• 原代码中cv2.flip(frame,0).tostring()在新版OpenCV中已弃用，替换为tobytes()避免运行警告
• 确保kv文件夹和Python主文件在同一目录下，kv文件路径和main.kv内的include路径完全匹配
• 新增的屏幕生命周期逻辑会在切出摄像头页面时自动释放摄像头、停止帧读取调度，避免后台空跑消耗性能

内容的提问来源于stack exchange，提问作者Cole Miller