Method, devices and arrangements for locating bony parts present in a poultry leg

An arrangement for training at least one neural network for locating bony parts present in a poultry leg, comprising:

• a plurality of poultry legs;
• an optical camera adapted to record images of front sides or back sides of the poultry legs in the optically visible wavelength range and configured to generate optical image data for each of the poultry legs;
• an X-ray source adapted to irradiate the back side or front side of the poultry legs with X-rays, and an X-ray imaging system adapted to record X-ray images on the side of the poultry legs that is remote from the X-ray source and configured to generate X-ray image data for each of the poultry legs;
• a display and input device configured to display the X-ray image data and/or to display hybrid image data and to input reference points which are to be defined and which serve to mark positions of the bony parts;
• an overlay unit configured to overlay the optical image data with the X-ray image data and/or the reference points in order to generate the hybrid image data for each of the poultry legs;
• at least one neural network;
• and
• a learning cycle control unit which is configured and adapted to input the optical image data as input data and the reference points as target data as training data for the neural network, wherein the learning cycle control unit is adapted to repetitively adjust weights of the neural network on a basis of a difference between the target data and the output data generated by the neural network.