Deep Learning Bundle

• A new code reader tool ECodeReader allowing to read multiple code types from a single integrated interface
• Deep Learning Studio is now available on Linux ARM-64 platforms
• Neural Network “engines” allow additional devices support (OpenVINO, Tensor RT) and increased speed.
• EasyBarCode2 now supports learning from images to improve its capabilities on similar images.

• A new assisted segmentation tool is available in Deep Learning Studio for the EasySegment Supervised annotations. It provides a quick and efficient way to get pixel- perfect annotations.
• EasyFind has been extended to support :
  • Vector learning patterns. This extension is well- suited to find objects with a known geometry.
  • Pattern matching using feature point technology
• EasyBarCode2 and EasyQRCode now support the GS1 symbology.
• EasyMatrixCode2 now supports the Data Matrix Rectangular Extension (ISO/IEC 21471, DMRE).

• EasyLocate Interest Point : New working mode (alternative to EasyLocate Axis Aligned Bounding Box) simplifying the annotation of the dataset and the configuration of the tools when all objects have the same size. A single click is enough to annotate an EasyLocate Interest Point object.
• EasyGauge : New Polygon Gauge in addition to Point, Line, Rectangle, Circle and Wedge gauges.
• Deep Learning tools now supporting GPU processing on NVIDIA Jetson.
• Faster compilation time with simplified C++ headers and reduced calling cost to the Open eVision API.

All Open eVision libraries are now also available for embedded ARM devices.

Deep Learning generally requires significant amounts of processing power, especially during the learning phase. Deep Learning Bundle supports standard CPUs and automatically detects Nvidia CUDA-compatible GPUs in the PC. Using a single GPU typically accelerates the learning and the processing phases by a factor of 100.

Neural Networks are computing systems inspired by the biological neural networks that constitute the human brain. Convolutional Neural Networks (CNN) are a class of deep, feed-forward artificial neural networks, most commonly applied to analyzing images. Deep Learning uses large CNNs to solve complex problems difficult or impossible to solve with so-called conventional computer vision algorithms. Deep Learning algorithms may be easier to use as they typically learn by example. They do not require the user to figure out how to classify or inspect parts. Instead, in an initial training phase, they learn just by being shown many images of the parts to be inspected. After successful training, they can be used to classify parts, or detect and segment defects.

EasyClassify is the classification tool of Deep Learning Bundle. EasyClassify requires the user to label training images, that is to tell which ones are good and which ones are bad, or which ones belong to which class. After this learning/training process, the EasyClassify library is able to classify images. For any given image, it returns a list of probabilities, showing the likelihood that the image belongs to each of the classes it has been taught. For example, if the process requires setting apart bad parts from good ones, EasyClassify returns whether each part is good or bad, and with what probability.

EasySegment is the segmentation tool of Deep Learning Bundle. EasySegment performs defect detection and segmentation. It identifies parts that contain defects, and precisely pinpoints where they are in the image. The supervised mode of EasySegment works by learning a model of what is a defect and what is a “good” part in an image. This is done by training with images annotated with the expected segmentation. Then, the tool can be used to detect and segment the defects in new images. The supervised mode of EasySegment achieves better precision and can segment more complex defects than the unsupervised mode thanks to the knowledge of the expected segmentation.

Open eVision includes the free Deep Learning Studio application. This application assists the user during the creation of the dataset as well as the training and testing of the deep learning tool. For EasySegment, Deep Learning Studio integrates an annotation tool and can transform prediction into ground truth annotation. It also allows to graphically configure the tool to fit performance requirements. For example, after training, one can choose a tradeoff between a better defect detection rate or a better good detection rate.

• Neo is the new Licensing System of Euresys. It is reliable, state-of-the-art, and is now available to store Open eVision and eGrabber licenses.
• Neo allows you to choose where to activate your licenses, either on a Neo Dongle or in a Neo Software Container. You buy a license, you decide later.
• Neo Dongles offer a sturdy hardware and provide the flexibility to be transferred from a computer to another.
• Neo Software Containers do not need any dedicated hardware, and instead are linked to the computer on which they have been activated.
• Neo ships with its own, dedicated, Neo License Manager, which comes in two flavours: an intuitive, easy to use, Graphical User Interface and a Command Line Interface that allows for easy automation of Neo licensing procedures.

• Deep Learning Bundle has been tailored, parametrized and optimized for analyzing images, particularly for machine vision applications.
• Deep Learning Bundle has a simple API and the user can benefit from the power of deep learning technologies with only a few lines of code.
• Try before you buy: Deep Learning Bundle comes with the free Deep Learning Studio training and evaluation application.

EasySegment is the segmentation tool of Deep Learning Bundle. EasySegment performs defect detection and segmentation. It identifies parts that contain defects, and precisely pinpoints where they are in the image. The unsupervised mode of EasySegment works by learning a model of what is a “good” sample (i.e. a sample without any defect). This is done by training it only with images of “good” samples. Then, the tool can be used to classify new images as good or defective and segment the defects from these images. By training only with images of good samples, the unsupervised mode of EasySegment is able to perform inspection even when the type of defect is not known beforehand or when defective samples are not readily available.

EasyLocate is the localization and identification library of Deep Learning Bundle. It is used to locate and identify objects, products, or defects in the image. It has the capability of distinguishing overlapping objects and, as such, EasyLocate is suitable for counting the number of object instances. Two methods are available:

• “EasyLocate Axis Aligned Bounding Box” predicts the bounding box surrounding each object (or defect) it has found in the image and assigns a class label to each of them. It must be trained with images where the objects (or defects) that must be found have been annotated with a bounding box and a class label.
• “EasyLocate Interest Point” predicts the position (as one point, typically the center, but may be otherwise defined) for each object (or defect) it has found in the image and assigns a class label to each of them. All the objects (or defects) in the image must have the same approximate size. It must be trained with images where the objects (or defects) that must be found have simply been annotated with an interest point and a class label. The annotation process is faster with EasyLocate Interest Point as a single click is enough to annotate an object.

• Windows 7 to Windows 10 x86 (32-bits) and x86-64 (64-bits)
• Windows 11 x86-64 (64-bits)
• Linux 64 bits (x86-64 and ARMv8-A) with a glibc version 2.18 or newer