Data Monsters introduce NVIDIA-based AI solution for quality control on high-speed conveyors.
It’s time to get rid of manual checks, thanks to the latest computer vision self-learning quality control by Data Monsters! In real-time, including high speed conveyors, the system automatically identifies a product that visually differs from norm, and alerts the workers.
Replace selective manual checks with fully-automated control:
It is a high-speed packaging line, filling up to 30 aluminum cans per second. Cans are covered with condensate and positioned on the line randomly. Package design of the cans may change from week to week.
Every 30 minutes, a worker manually takes 2 cans from the line and checks them. If a single defect is detected, the line must be stopped, all the produced cars are marked as blocked and are subsequently dumped.
The amount of dumped product is estimated to be millions of cans per year.
Data Monsters trained a real-time pipeline that identifies visual anomalies on the product, and alerts the packaging line operator in real time.
Our client produces life-critical medical devices with batteries. Therefore, all the batteries pass a quality check using x-ray machines, and operators then manually check every image. This is a slow, time-consuming, and unreliable process.
Data Monsters trained a real-time pipeline that detects anomalies by measuring and segmenting the x-ray scans.
Our client produces equipment that analyzes and finds defects in semiconductor wafers. The challenge is to identify new defect patterns that may indicate new problems with the manufacturing process or equipment.
Data Monsters trained the computer vision pipeline to identify anomalies and different defect patterns.
Our processes are 100% aligned with the strategy of our partners. If you need a reliable R&D partner - Data Monsters is a great and scalable contractor and AI solutions supplier.
Get personalized advice from Data Monsters who've spent more than 10 years on building AI solutions. Our whitepapers, our case studies, including "7 reasons why AI may fail in your project. How to make it work." whitepaper.