摘要 |
A hybrid machine learning system (100, 200, 300, 400, 500, 600, 700, 800) is for processing image data (110, 210, 310, 410, 510, 610, 710, 810) obtained from an image sensor (105, 205, 305, 405, 505, 605, 705, 805). The system comprises a front end (115, 215, 315, 415, 515, 615, 715, 815) comprising one or more hard-coded filters (120, 220, 320, 420, 520, 620, 720, 820). Each of the one or more hard-coded filters is arranged to perform a set task. The system also comprises a neural network (125, 225, 325, 425, 525, 625, 725, 825) arranged to receive and process output from the front end. The one or more hard-coded filters include one or more hard-coded noise compensation filters that are hard-coded to compensate for a noise profile of the image sensor from which the image data is obtained. |