摘要 |
<p>Disclosed are a detection method and system for a locally deformable object based on on-line learning. The method comprises: step 1, using a sample image in a training set to train a locally deformable object detection model so as to obtain the locally deformable object detection model after initial training; and step 2, using the locally deformable object detection model to perform object detection on an image to be detected, and using a method for a GUI labelling on-line learning to update and optimize the existing locally deformable object detection model. The above-mentioned method of the present invention distributes the whole time-consuming training process in each object detection; at the same time, the model can be updated in real time, robustness of the locally deformable detection model would be further improved, and the demand for memory is not much. According to the above-mentioned method of the present invention, under the background of big data, data used for object detection can be quickly and effectively processed.</p> |