Neurocle lowers inspection cost and enables fast deployment with auto deep learning technology.

Reducing the defect rate of products has long been a concern of all manufacturing companies. For this purpose, the vision inspection of the product has been developed from human to machine. With the development of technology, companies that apply deep learning vision technology to perform vision inspection are increasing for reducing the defect rate.

However, companies that have introduced or are considering adopting deep learning vision technology are complaining of various practical difficulties such as △Difficulty of data management △The burden of cost hiring deep learning researchers and advanced developers △High-end hardware

Accordingly, Hongsuk Lee(CEO of Neurocle Inc.) introduced Auto Deep Learning Vision Inspection at 2020 BUSAN·ULSAN·GYEONGNAM SMART FACTORY CONFERENCE & EXPO and said “Now, everyone can use the optimal deep learning model without a deep learning expert.”

At this conference, Hongsuk Lee(CEO of Neurocle Inc.) gave a lecture on the subject of 'Application Cases of Auto Deep Learning Algorithm and Solution Overcoming the Limits of Existing Vision Inspection'.

Neurocle's Auto Deep Learning vision inspection can be applied to various platforms and enable data management and auto-optimized modeling and applied to various tasks easily. Users can analyze images by pixel unit to detect the defective area of the product, and the product can be classified by good/bad or by classes. Also, it is possible to identify the location or number of products in the image, and even the text in the image, such as the product serial number, can be identified. In addition, it is possible to detect defective products by learning only normal images.

Hongsuk Lee(CEO of Neurocle Inc.) said, “Especially in the manufacturing field where a lot of time and manpower are spent on quality inspection for inspection, auto deep learning vision inspection is efficient way.”

Neurocle's Auto Deep Learning Vision Inspection Solution is currently being applied to automotive parts manufacturers' 20-class defect detection processes and cosmetics logistics inspection processes of logistics companies. It is also used for endoscopic image analysis and microscopic cell image analysis.

Source :