EasySegment deep learning library

Share this on social media:

Euresys is proud to announce the availability of a new Deep Learning library: EasySegment.

EasySegment works in unsupervised mode. After being trained with “good” images only, it can detect and segment anomalies and defects in new images. EasySegment works with any image resolution, supports data augmentation and masks, and is compatible with CPU and GPU processing.

The new EasySegment complements EasyClassify as Open eVision's Deep Learning libraries. EasyClassify is a classification tool that can detect defective products and sort them into various classes.

EasyClassify and EasySegment are easy to use and have been tailored, parametrized and optimized for analyzing images, particularly for machine vision applications.

They come with the free Deep Learning Studio application, which allows users to evaluate the capabilities of the libraries, create and manage datasets and supervise the training process.

EasySegment - Deep Learning segmentation library

- Unsupervised mode: train only with “good” images to detect and segment anomalies and defects in new images
- Works with any image resolution
- Supports data augmentation and masks
- Compatible with CPU and GPU processing
- Includes the free Deep Learning Studio application for dataset creation, training and evaluation
- Only available as part of the Deep Learning Bundle

EasyClassify - Deep Learning classification library

- Includes functions for classifier training and image classification
- Able to detect defective products or sort products into various classes
- Supports data augmentation, works with as few as one hundred training images per class
- Compatible with CPU and GPU processing
- Includes the free Deep Learning Studio application for dataset creation, training and evaluation
- Only available as part of the Deep Learning Bundle

The object detection tool can locate trained object classes and separate those touching or partially overlapping.

27 April 2023

With the Deep Counting feature, large numbers of objects such as glass bottles can be counted.

19 April 2023

The new diffractive optical network-based multispectral imager achieves high imaging quality and high spectral signal contrast. (Image: UCLA)

07 April 2023