Neurocle is an innovative deep learning vision software company that enables the application of deep learning technology in all industries. The company’s deep learning vision software interprets images and video data by integrating deep learning technology with computer vision. The main strength of Neurocle’s software is the GUI-based Auto Deep Learning Algorithm that allows even those without deep learning expertise to create the best-performing deep learning models easily.
Thanks to the hard work of their members, Neurocle was chosen for the Innovators Awards by Vision Systems Design for two years in a row and was named an AI Computer Vision Cool Vendor by Gartner. The company continues to aim to provide easy software that anyone can use to solve all kinds of computer vision problems with deep learning.
Easy Engineering: What are the main areas of activity of the company?
NEUROCLE: Neurocle works with industries where deep learning could result in better inspection standards than conventional rule-based or visual inspections. Currently, the most active sector is electronics. Our software is widely adopted by various inspection processes in the electronics industry, such as rechargeable battery, PCB, and semiconductor production, where high precision and accurate inspection are required.
We are also in the process of expanding our presence in the global market. We already have a strong partner network in Asian markets, including Korea, Japan, China, Singapore, Malaysia, and Taiwan. In European markets, we have initiated a partnership with an esteemed system integration company in the Netherlands and have been actively discussing with a few companies in other European countries.
E.E: What’s the news about new products?
NEUROCLE: This October, Neurocle plans to release the latest version 3.1 software packages. Along with minor bug fixes and feature enhancements, the most significant update is the addition of the Rotation Correction model. Previously, we had five models: Classification, Segmentation, Object Detection, Anomaly Detection, and Optical Character Recognition (OCR). With the new Rotation Correction model, users can now skip the process of manually rotating images in the pre-processing stage and take advantage of the simplified process of making images upright.
Also, this update package includes a new functionality of Inference Center. Inference center provides flowchart function by connecting deep learning models in the order a user want to inference. Users can now find the inference result from their existing deep learning models and a new inference result of additional images without the help of vision engineers.
E.E: What are the ranges of products?
NEUROCLE: Neurocle provides three deep learning software packages: Neuro-T Auto Deep Learning Vision Trainer, Neuro-X Deep Learning Model Trainer for Experts, and Neuro-R Runtime API Engine.
– Neuro-T is no-code deep learning vision software for people with little understanding of coding and deep learning.
– Neuro-X is a deep learning model trainer for deep learning experts. Users can create optimized models by adjusting hyperparameters, allowing greater control and autonomy.
– Neuro-R is runtime software that helps model deployment and supports optimized integration with various inference platforms, including GPU and embedded processors.
E.E: At what stage is the market where you are currently active?
NEUROCLE: The machine vision market is rapidly growing but is still in its early stages. Vision technology is already being used in many fields, such as CMOS sensors, embedded vision, deep learning, robot interface, data transmission standardization, and image processing. More than just improving performance in production lines, the market has now developed into focusing on the integration and automation of deep learning technology.
Since the introduction of deep learning technology in machine vision systems, the importance and impact of deep learning have been widely recognized. However, the mass deployment of this technology has yet to be popularized. Neurocle hopes to lead innovation at this stage and be the major player in the market by making deep learning accessible to the masses.
E.E: What can you tell us about market trends?
NEUROCLE: Conventional machine vision systems only promise stable performance for uniform, structured products. But as the list of exceptions and defect types grow longer, proper inspection becomes more difficult. Slight changes in parts due to magnification, rotation, placement, and distortion can be inspected with the conventional machine vision system, but complex surface texture and image quality problems require more complicated systems. Defects that affect the function of parts can be detected but assessing variability and deviation between visually similar characteristics is problematic with existing machine vision systems.
Deep learning image analysis provides a solution to this issue, introducing opportunities for factory automation in all industries. Image analysis based on deep learning combines the sophistication and flexibility of visual inspection with the reliability, consistency, and speed of computer systems. This can address the challenges of vision applications, where traditional machine approaches have failed. As a result, more and more companies are adopting deep learning-based solutions.
E.E: What are the most innovative products marketed?
NEUROCLE: Although the obvious goal is to provide high-performing deep learning models, we believe that true innovation is in this concept: no-code AI platforms enabling businesses of all kinds to benefit from powerful technology that has previously only been accessible to large organizations.
Neuro-T is a no-code deep learning vision software that automatically optimizes AI architectures and hyper-parameters. Anyone can create and optimize the best performing deep learning vision models with just a few clicks with Neuro-T. Moreover, features like Flow Chart, Auto-Labeling, and Fast Retraining makes it easy for users to design projects, label images, and maintain models.
E.E: What estimations do you have for the rest of 2022?
NEUROCLE: We are very confident in the value of our products in the market, and we are looking for more partnerships with qualified system integration partners worldwide.
This fall, Neurocle will exhibit the latest software packages and demonstrate software applications at Vision Stuttgart in Germany (Oct 4-6) and at the Vision Show in Boston (Oct 11-13). More information can be found on our website www.neuro-cle.com.