EasySegment, a new addition to Euresys’s deep learning software library, enables deep learning systems to work in unsupervised mode, trained only with “good” images, i.e. images that meet inspection criteria, which then enables the deep learning system to determine “bad” images when detecting and segmenting anomalies in new images.
Any image resolution can be used by EasySegment. The software also supports data augmentation and masks, is compatible with CPU and GPU processing. EasySegment is designed to complement EasyClassify, another tool in the Euresys deep learning software library, that detects defective products and sorts them into classes.
EasySegment is only available as part of the Euresys deep learning bundle and includes the company’s Deep Learning Studio application.
To Learn More:
Headquarters: Angleur, Belgium
Product: EasySegment deep learning library
Key Features: Learns from “good” images only, supports CPU and GPU processing, supports data augmentation and masks.
What Euresys says: View more information on the EasySegment deep learning library.