Image segmentation
Using the method of neural network to extract the mask of the target to be measured in the image, the target of the same class will be assigned the same mask pixel value.
Register for classification
Classify images according to registered image categories.
Instance segmentation
The method of extracting the mask of the target to be measured in the image by using the neural network can accurately output the mask information of each target instance.
Image retrieval
A technique that uses image semantic information (such as color, texture, layout, etc.) for analysis and retrieval.
Unsupervised segmentation
Unsupervised segmentation is an unsupervised mode, and there is no need to label defective samples, only OK samples need to be trained.
Anomaly detection
It is suitable for the situation where the target shape is fixed, the defect type is fixed, and there are many OK samples and few NG samples.
Image classification
The neural network is used to extract image features and give category information according to the distribution of their features.
Object detection
A method of extracting the position information of the target to be measured in an image by using a neural network.
