EasySpotDetector
(Discontinued)
Advanced Surface Inspection For Battery, Paper, Film, Or Glass Industries.
- Detection of faint defects and contamination, even in noisy images
- Fast processing for in-line inspection
- Compatible with acquisition from line-scan and 2D cameras
- Optional pre-alignment of the region of interest on part’s edges
- Optional Deep Learning classification of the defects
- Simple and comprehensive C++, C# and Python API
Description
Realtime Processing For In-Line Surface Inspection
With its two-stage approach, EasySpotDetector is faster than other Deep Learning based object segmentation processing. EasySpotDetector can process up to 200 MPixels per second on an Intel i7-10850H computer (detection only). The classification can benefit from a GPU operation but is also optimized to run on CPU thanks to OpenVINO.
Set Of Parameters To Control Defect Segmentation
A set of explicit parameters allows the user to target specific defects. The type (particle, scratch…), the aspect (lighter, darker or both), the size, the minimum contrast (strong or faint defect) of the defects can be adjusted.
Simple And Comprehensive API
EasySpotDetector provides a single API for the alignment of the region of interest (ROI), the detection of defects on surfaces and the classification with a custom trained Deep Learning classifier.
Custom Trained Deep Learning Object Classifier
Detected objects can be submitted to a deep learning classifier. The classifier is trained by the user, specifically for his particular application, using the user-friendly Deep Learning Studio. Possible usages of the classifier are:
- Confirm or invalidate the detected candidates.
- Evaluate the severity level of the defects.
- Split the detected objects into several classes based on their aspect.
Tested On Various Use Cases
EasySpotDetector has been successfully tested on several surface inspection applications, including: battery foil, fabric, steel, passive electronic components and natural materials e.g.: leather, wood.
EasySpotDetector Illustrations
New Open eVision Studio
Complex image processing sequences can be designed using a graphical interface. The collection of tools represents the diversity and the capabilities of the Open eVision libraries. The C++, Python and C# source code, corresponding to the processing pipeline, is automatically generated and provides an interactive documentation of the Open eVision API. The New Open eVision Studio can process live image sources such as a GigE Vision camera, a Coaxlink frame grabber or eGrabber recorder sequences.
This application is free of charge, runs on Windows, Linux and is compatible with Intel and ARM 64-bit architectures.
