Dongsheng (Suzhou) Intelligent Technology Co., Ltd.

AI Camera

The deep learning intelligent camera created by Dongsheng Intelligent Technology is a highly integrated micro machine vision system, with an integrated structure that is flexible and easy to deploy. The camera perfectly matches with the platform of Handdle Al algorithm and supports the import of Al models, which saves 90% of Al deployment time and effectively solves complex scenario problems in various industrial fields.

DS-ATS200 Series

The DS-ATS200 Series AI camera jointly released by Dongsheng Intelligent Technology and Huawei Ascend is equipped with deep learning algorithms, which can provide edge AI computing capabilities on the basis of the Huawei Ascend Atlas 200 AI acceleration module and handle complex defective scenarios efficiently and accurately.

DS-NX Series

The DS-NX Series AI camera independently developed by Dongsheng Intelligent Technology is equipped with deep learning algorithms and embedded with NVIDIA accelerator cards, which is mainly used for flaw detection, OCR character recognition, visual positioning and other industrial intelligent manufacturing scenarios under complex backgrounds.

Size measurement

ROI segmentation

Code reading

Feature matching

Template matching

OCR

Object detection

Defect classification

Core Functions

Defect segmentation

Target location

Performance edges

Faster detection

It can optimize the image preprocessing, neural networks and underlying algorithms. On the basis of ensuring the detection accuracy and precision, it can detect small articles much faster than traditional algorithms.

Lower dependence on data

It can reduce the dependence of Al algorithm on the size of data by the small-sample training technology, realize quick training and deployment with very little data, and thus own high accuracy.

More universal model

It can improve the detection of the neural network models by the transfer learning technology, so that the model can be transferred to more product models and solve the application scenarios of fewer samples and more models.

Higher detection accuracy

It can greatly reduce the over-detection rate and improve the accuracy by the application of technology in enhancing the image, extracting the multi-dimensional feature, identifying the feature accuracy and reinforcing the detection of pixel coincidence.