AI Classification (Multi-Class)

This Step is used to perform intelligent classification of target objects in images into multiple classes. Through image acquisition, ROI selection, and class labeling, the system uses deep learning for training and validation to automatically identify each object’s specific class.

This Step is suitable for scenarios requiring accurate classification of objects. This Step typically follows the 2D Smart Camera Step, or Steps such as 2D Matching. The former is suitable for scenarios where the target position is relatively fixed, performing classification judgment after acquiring image data; the latter is suitable for scenarios where the target position may shift, using alignment Steps to obtain the target’s pose transformation parameters, which are input to this Step to synchronously adjust the ROI position for stable classification judgment.