PEKAT VISION learns to understand the product from a set of images and is able to find anomalies, detect and classify defects, and check the integrity of surfaces. It is well suited for inspection of textile and other woven and knitted fabrics, non-woven fabrics, leather, plastic or rubber foils, filter fillings, paper, surgical sheets and many other materials.
Using the Anomaly Detector module, defect of previously unknown shape, pattern, and size are identified. Furthermore, the results are not diminished by an irregular patters on the fabric as it learns and recognizes the pattern, something that may be difficult with manual inspection. For clarity, the defect is always highlighted with a heat map.
Textile Inspection
The Anomaly of Surface module is a good choice for detecting defects that are hard to predict or when we don’t know what they could look like. In this case PEKAT VISION is trained on defect free images only
Raw Leather Inspection
Using PEKAT VISION, defects in the leather are highlighted with a heat map. Early identification of defects and irregularities, before processing the material further, is an effective way to eliminate or minimize waste
Inspection of Leather for Upholstery
Defects in pattern on leather for a high-end car upholstery caused by a damaged tool are identified early with PEKAT VISION Anomaly Detector module
Sack Fabric Defect Detection
The Anomaly Detector module also has its use in the inspection of fabrics used in the production of agricultural bags
Non-woven Fabric Inspection
Even when batches are changed often and the exact size or shape of a possible defect is not known, the Anomaly Detector module still achieves a very high accuracy level
Anomaly Detection of Denim Fabric
Thanks to its unsupervised training, PEKAT VISION Anomaly Detector module, achieves accurate results with minimum effort and can be used for a vast variety of textile fabrics including denim fabric
And more! Ask for a free feasibility study for your specific case. Our software is highly versatile and can be customized to suit nearly any intended application.