Generating training data for identifying dishes

A method for providing training data for identifying a dish, the method comprising the steps of: detecting an image of the dish during a disposition of the dish in a cooking apparatus, where the image is detected from a predetermined perspective; determining an object boundary of the dish on the image; and providing training data including the image and the determined object boundary. 一种用于提供训练数据用于识别菜肴的方法，该方法包括以下步骤：在菜肴布置在烹饪设备中期间，检测该菜肴的图像，其中，从预确定的视角检测该图像；在该图像上确定该菜肴的对象边界；和提供包括该图像和所确定的对象边界的训练数据。