News

Technology News 26 October, 2021

Neurocle is a deep learning vision software company that hopes to enrich the world with their deep learning technology, striving to catalyse success for clients and partners to embark on an innovative journey with their releases of products -Neuro-T and Neuro-R.

Guide to using Neurocle’s Deep Learning Vision Software -Neuro-T and Neuro-R.

Anyone can create and utilize deep learning vision models with Neuro-T and Neuro-R. Neuro-T automatically trains images and builds them a deep learning model. Neuro-R enables users to run real-time image analysis across a variety of areas including manufacturing, logistics and medical field with models customised by users themselves.

Creating a deep learning vision model, in Neuro-T, first upload your training images and go to the data tab, select the model data type that best fits your image analysis objectives. User can create multiple types of models with one data set. The model types include classification, segmentation, object detection, OCR and anomaly detection.  Classification refers to sorting of images to different categories.  Segmentation is recognising the shapes and locations of objects within images. Objection Detection, distinguishing the class and location of objects within images. OCR detects and extracts characters from images. Lastly, Anomaly Detection, identifies outliers that deviate from the majority of the data.

If there is an area that needs to be focused or excluded, for an accurate image training, specify it and designate a class to each image, splitting the images into training and model performance testing sets. In the train tab, check the label information, adjust the settings, and start training.

Neurocle’s auto deep learning algorithm automatically optimises the images’ parameters. Check the created model in the evaluation tab, user can visualise the model performance by checking the test images as well as writing a report about the results.

Finally, export deep learning model as a .net file. By using the information from the report, the user can write a research paper and build a smart factory by integrating deep learning models into his equipment using Neuro-R. -Apply a generated deep learning vision model for visual inspection.

Create the utilized optimal deep learning models with Neuro-T and Neuro-R with the automatic deep learning algorithm, making deep learning vision technology more accessible.

Unique Features of Neuro-T and Neuro-R

Neuro-T and Neuro-R includes unique features such as 1. Auto Deep Learning Algorithm, 2. Fast Retraining Feature, 3. Server-Client Based Data Management System, 4. Fast Inference Time and 5. Supports Diverse Platforms.

1. Auto Deep Learning Algorithm

To create optimized deep learning models in one simple train. Through Auto Deep Learning Algorithm, even non-experts can easily create high-performance models and save valuable time and resources compared to traditional models.

Generally, models require a long lengthy process to complete and reach the final model stage. They must go through DL Architecture to Hyperparameter to Data Augmentation and repeated trainings before arriving at the final model. However, with Neuro-T, a simple click of auto optimization through auto deep learning can achieve a high-performance model.

2. Fast Retraining Feature

Guarantees performance and maintenance of existing DL models. Through Fast Retraining, models can be more quickly and accurately adapted to changes in inspection equipment or environment than general retraining or transfer learning methods.

General retraining requires several trails and prolonged time and collection of all train data and even prolonged training time. Whereas, transfer learning, additional data to be collected and short training time does not guarantee high performance due to the lack of data. With Neuro-T, all the collected train data adding on short training time can achieve high accuracy in short training time, fast retraining.

3. Server-Cilent Based Data Management System

Collaborate with a single license. Neuro-T’s server-client architecture makes it possible to effectively collaborate with users without physical, spatial and device constraints. Comparing other general project management structures to Neuro-T’s project management structure, general project management structures consist of high device dependency and requires purchasing a license for each device which means local PCs cannot be connected to other local PCs with the same license. However, Neuro-T’s project management structure is possible to collaborate by accessing to one software from multiple devices. This means that a local server PC can be connected to not only another local PC but as well as Client PC, Client Laptop and Client Tablet PC, bringing way more convenience and accessibility compared to general project management structures.

4. Fast Inference Time

Provides high inference speed for sites where processing time is crucial. Inference speed can be freely adjusted as needed and enables fast inference speed to real-time inspection. The inference time at level one is specialised for fast speed which may be faster during batch processing. For example, 1.4ms based on Object Detection 8 Batch. At Level 1, classification processing time is 1.3 to 1.7ms, segmentation is 2.2 to 2.3ms and object detection is 3.5 to 4.0ms. If level 1’s speed is too fast, level 3 does not have any speed limit. For classification, processing time at level 3 is 1.3 to 3.8ms, segmentation is 2.2 to 4.5ms and object detection is 3.5 to 6.8ms.

5. Supports Diverse Platforms

Customise to any device regardless of its specification. Using Neuro-R, apply your model to any platform such as embedded processors, CPUs, and GPUs that best suits your industry. Neuro-R can be used on various platforms ranges from light industries such as consumers good crops with platforms in embedded-based like jetson to heavy industries such as manufacturing, for example semiconductor, display and security image using PC-based multi-GPU.

Technology News 3 September, 2021

Canon Quality for Industrial, Machine Vision & Scientific Applications

Canon’s CMOS sensor technology offers OEMs and end users expanded possibilities for industrial vision applications. From uniquely large pixel sizes to exceptionally high pixel counts in a compact design, Canon CMOS sensors push the boundaries of imaging possibilities. Facilitating on-going innovation in industrial camera and image capture technologies, Canon CMOS sensors support pioneering solutions for advanced industrial, machine vision, medical and scientific applications.

A Top-of-the-Line CMOS Sensor

With rolling shutter and a sensor size of 7.9 x 4.5 mm, the Canon LI7050SAC is a top-of-the-line CMOS Sensor. It offers a dynamic range of 120 dB and a maximum frame rate of 60 fps, creating flexibility and performance. It sports RGB color options and a sensitivity of 54,000, making it ideal for a number of applications. The Canon LI7050SAC CMOS Sensor offers 2M pixel HDR with high sensitivity at an affordable price. It’s square 4.1µm x 4.1µm pixel arrangement and 1.8” frame achieves ultra-high resolution in a compact design. Discover how you can take your machine vision project to new heights with the LI7050SAC from Canon.

Kindly contact us for further information. Email us info@jm-vistec.com

Technology News 3 September, 2021

Neuro-T
Our deep learning vision software trains models for image recognition. Its intuitive GUI and auto deep learning technology makes it possible for non-experts to create high-quality models with just a few clicks.

Neuro-R
Our deep learning vision software utilized to run models.
Models are deployed and embedded to possibly any device.

Neurocle is a company specializing in deep learning-based vision software, has released 2.3 version of deep learning deep learning vision software Neuro-T & Neuro-R, which includes a variety of new features.
Neuro-T and Neuro-R are deep learning vision software that can be used by non-experts. It is provided in an intuitive graphical UI (GUI)-based design method so that even non-experts can utilize deep learning vision technology. Users are allowed to design processes that automate appearance defect inspection, medical image analysis, and logistics packaging inspection.

What is worthy of noticing about the newly released version 2.3 of Neuro-T & Neuro-R is that Fast Retraining function that enables rapid maintenance/repair is added. There were many customers confronted with difficulty of maintenance/repair of the inspection process due to the addition of defective types or quick change of the product in the production process. The majority of these customers tried to solve this problem through Transfer Learning. Transfer learning is a method of learning images by adding data to existing models, and often does not clearly reflect new changes due to the many constraints of existing models. However, the Fast Retraining function is a learning method that utilizes existing architectures and hyperparameters when learning data is added or changed, allowing data to be re-learned across the board, leading to clear and fast maintenance. Neurocle also present a method to increase the accuracy of image learning by adding pre-processing features to set up ROI(Region of Interest) and MASK(area excluded), as well as reflect improvements required by customers in previous versions, such as adding function to remove data augmentation, improving Labeling Tab UI, and adding shortcuts.

Neurocle CEO Hongsuk Lee said, “Neurocle has been providing customers with the process of automatically generating models by utilizing auto deep learning algorithms since the early version, thereby securing the convenience of learning.” “We wanted to further maximize the efficiency and economy of the deep learning project by adding fast retraining function and image pre-processing, considering the situation of relearning due to changes in learning data,” he said. Meanwhile, Neurocle has recently succeeded in registering a number of core technology patents based on innovative products, and has successfully signed contracts with leading conglomerates and university hospitals, posting sales growth of more than 200% annually. In addition, it has been recognized at home and abroad, winning the Korea Impact Tech Award hosted by the Ministry of Science and ICT and the Innovators Award hosted by Vision System Design.

Technology News 18 August, 2021

More resolution! More image quality!
With high-resolution machine vision systems, the smallest object structures are classified in the highest quality. However, all sensors available on the market with resolutions with 100 megapixel and above only use a rolling shutter. The new shr661 CMOS camera from SVS-Vistek can confidently be called a milestone in machine vision, because it offers an enormous resolution of 127 megapixels and, despite this extremely high resolution, works with a global shutter. It thus enables new applications in numerous fields of use.

• Global Shutter – why?
  CMOS camera sensors can be built in two different ways: With rolling shutter or with global shutter. With global shutter, in simple terms, all pixels of the sensor are exposed simultaneously and then read out. With rolling shutter, the sensor lines (overlapping) are exposed and read out one after the other. Rolling shutter sensors are technically easier to implement, which is why highest resolution cameras typically use this shutter variant. However, the time-shifted exposure of the lines in the rolling shutter results in different exposure times in the image field, which can lead to distortions and lighting-related pattern artifacts. Therefore, in applications with moving objects, the use of such sensors is very limited. In most machine vision applications, global shutter sensors are therefore used wherever possible. The shr661 is the first camera in this resolution class that can inspect moving objects (e.g. test specimens on a conveyor belt).
• First on the market
  As a recognized specialist for high-resolution industrial cameras, SVS-Vistek is the first manufacturer worldwide to show a camera with 127 megapixels and global shutter – the shr661. It is based on the Sony Pregius sensor IMX661 with a pixel size of 3.45 µm edge length. With this camera, all those tasks where a global shutter is helpful or even necessary are suddenly possible, even in the highest resolution. The image quality of the shr661 is the current high-end of industrial cameras.
• Industry friendly
  The design, which is consistently geared towards industrial use, facilitates integration into industrial applications. A wide temperature range in operation is made possible by SVS-Vistek’s proven, highly effective cooling system. The excellent thermal management ensures an even temperature distribution on the sensor and excellent image homogeneity with outstanding dynamic range. Despite the large sensor, the camera has a small, integration-friendly housing. The 3.45 µm pixels are not critical when choosing suitable lenses. And the high resolution of 13392 x 9528 pixels, together with the global shutter, reproduces the finest structures even with moving objects.
• State of the art interfaces and high end features
  The high data rate of the huge sensor of up to 2.2 GB/s requires modern, fast interfaces. By using the powerful CoaXPress-12 interface, the shr661 transmits up to 17 fps and in the version with 10GigE interface 6.5 fps are achievable.
  Industrial cameras with the highest resolutions, such as SVS-Vistek’s shr661, benefit especially from Flat Field Correction and Defect Pixel Correction, which can also be adapted by the user to the respective scenario. The shr661 can be operated with 8, 10 or 14 bit colour depth. In all colour depths, the outstanding homogeneity of the image is worth emphasizing.
• Easy to integrate
  As you would expect from SVS-Vistek cameras, integration of such a high performance camera is very simple. The I/O framework with multiple inputs and four power-out outputs supports precise triggering as well as sequencers and logical operators. The power-out outputs with the built-in strobe controller can drive LED lighting directly. The GenICam interface with GenTL interface enables easy integration into all major machine vision software packages within only a few minutes. SVS-Vistek’s shr661 is currently available as a demonstrator.

Technology News 1 December, 2020

8 Mega Pixel Visible ～ NIR Chromatic aberration corrected lens

Characteristic

• f12mm, 16mm, 25mm, 35mm, 50mm, 75mm are available for Fixed Focal Lens
• Super high resolution, Low distortion, and excellent uniformity of brightness
• Realizes color reproduction in color images
• 400nm～950nm、Focus Shift correction
• f16～50mm is available for Varifocal Lens

For more information, kindly send in your enquiries to info@jm-vistec.com or call +65 6748 5517.

Technology News 2 November, 2020

36x motorized zoom lens for 2 mega pixel

Characteristic

• F1.5 bright, suitable for the surveillance at night
• Zoom lens for 2 Mega Pixel
• Stable performance from wide to tele end

For more information, kindly send in your enquiries to info@jm-vistec.com or call +65 6748 5517.

Technology News 27 October, 2020

 SVS Vistek FXO Series

Characteristic

• Extremely smooth image
• Latest interface technology
• Powerful feature set

The FXO is the latest camera series from SVS-Vistek. The focus is on a high-end camera series in C-mount format with the latest sensor and interface technology. The FXO series starts with the new Sony Pregius Gen 4 sensors IMX540, IMX541 and IMX542.

For more information, kindly send in your enquiries to info@jm-vistec.com or call +65 6748 5517.

Technology News 19 October, 2020

Canon announced today the launch in Japan of the LI8020SAC (color) and LI8020SAM (monochrome), two new APS-H size (approximately 29.4 mm x 18.9 mm) CMOS sensors capable of image capture with a resolution of approximately 250 megapixels (19,568 x 12,588 pixels).

The new CMOS sensors achieve an ultra-high resolution of approximately 250 megapixels, enabling the image capture of wide areas while also preserving fine details throughout the image. With a pixel pitch of 1.5 μm (micrometers), a total of approximately 250 million pixels can be accommodated on an APS-H size sensor, making the sensors suitable for a wide variety of applications. From inspections of flat-panel displays (FPD), which continue to require increased precision due to the growth of 4K and 8K technology, to industrial inspections, video productions and digital archiving, wide-area monitoring and microscopes, the potential uses for the new sensors help meet a wide range of user needs.

The LI8020SAC and LI8020SAM realize image capture with an ultra-high resolution of 250 megapixels—approximately 125 times that of full HD (1920 x 1080 pixels) and approximately 30 times that of 4K (3840 x 2160 pixels)—allowing users to crop selected areas and digitally zoom to expand images while still retaining sufficiently high resolution.

As CMOS sensors increase in pixel count, the amount of signal also increases, leading to such problems as signal delay and slight discrepancies in timing. However, the new sensors, while delivering an incredibly high pixel count of approximately 250 million, features miniaturized circuitry and enhanced signal processing technology to achieve an ultra-high pixel readout speed of approximately 1,250 megapixels per second. This enables the sensors to realize ultra-high resolution image capture at up to approximately 5 frames per second (FPS), even during full pixel readout.

Technology News 16 October, 2020

10x Motorized Zoom Lens for 5～9 Mega Pixel

Characteristic

• Design for 1″
• Zoom lens for 5～9 Mega Pixel
• 3 motor (zoom, focus, and iris) type is also available, suitable for both industrial and surveillance

10x Motorized Zoom Lens for 5 Mega Pixel

Characteristic

• Design for 2/3″
• Zoom lens for 5 Mega Pixel
• Suitable for applications, required for high sensitivity for large format sensor

For more information, kindly send in your enquiries to info@jm-vistec.com or call +65 6748 5517.

Event Highlights,Technology News 13 October, 2020

For more information, kindly send in your enquiries to info@jm-vistec.com or call +65 6748 5517.