Articles

New Automatic Parameter Tuning feature in SPECTRO

Customers can now try the beta version available in SPECTRO Cloud SPECTRO’s flexibility to deal with different inspection objects comes from the ability to tune inspection model parameters to determine the way that SPECTRO looks at the target and how strict or tolerant it treats detected anomalies or defects. Finding the right parameters can be a time consuming task, but with our new auto-tuning feature SPECTRO can autonomously help customers find suitable parameters for an inspection task - significantly reducing the…

Hyperspectral Imaging and Sparse Modeling – Practical Experiments

This is Ryuji Masui, Data Scientist at HACARUS. In this article, I'm going to focus on how to apply Sparse Modeling to Hyperspectral Imaging, a technology HACARUS has strong interest in. Hyperspectral Image Let's start off by explaining what Hyperspectral Images (HSI) are. Normal cameras that are familiar to us are called RGB cameras, based literally on the three primary colors of light: red, green, and blue. Such cameras sense each wavelength to capture images. In simpler terms, by mixing…

Digital Imaging and Communications in Medicine(DICOM) Annotation using Computer Vision Annotation Tools(CVAT)

TL;DR - CVAT was used as an annotation tool for medical AI development - CVAT was adjusted to be compatible to read DICOM - A class to read in DICOM was added to cvat/apps/engine/media_extractors.py - Some features that will be useful in real projects, such as default label settings were also added 1. About CVAT Hello, my name is Yuya Sumie, an intern at HACARUS. In order to develop machine learning algorithms, we need to know how to finalize the…

Human Pose Estimation Experiment

Today's blog will be delivered by Yoshinobu Ogura, Data Scientist at HACARUS. Hello, my name is Yoshinobu Ogura, and I am a HACARUS Data Scientist. In this article, I will be using deep learning to estimate human skeletal structure - known as "Human Pose Estimation". About Human Pose Estimation The final goal aim of this article is to create a baseline for Human Pose Estimation. Human Pose Estimation is a technique to model the human skeleton from images, which is…

AI Operating On CPU

SPECTRO CORE

A toolkit for your EDGE visual inspection use cases. With a focus on execution speed at the EDGE, we have created a native  library supporting SPECTRO's main inspection algorithms. The ability to integrate SPECTRO's inspection features with industrial equipment has been a frequent request by our customers - today we are excited to announce that this feature is now available! CPU only – no GPU required By using SPECTRO CORE, system integrators or machine manufacturers can easily execute SPECTRO’s core inspection…

Advanced Applications of Sparse Modeling for Image Processing – Deficit Interpolation, Anomaly Detection, and Super-Resolution

This article originally appeared in Japanese on Codezine, below follows HACARUS translation of the same: This series of articles is for engineers who want to jump into machine learning and who have experience with deep learning and machine learning. The focus will be on a technique called "sparse modeling" - where you can perform analysis without large amounts of data. In the last article, image reconstruction using dictionary learning was covered. This time, three advanced methods will be introduced: deficit…

Sparse Modeling Study Sessions: A Convex Optimization Approach to LASSO – What is the Convex Optimization Problem?

In this article, Takahiro Suzuki, a data scientist at HACARUS explains the convex optimization problem covering all the basic terms you need to understand it. 0. Introduction My name is Takahiro Suzuki, data scientist in the HACARUS Tohoku team. As mentioned in a previous blog,  HACARUS has started study sessions to learn more about sparse modeling as part of our everyday business. The Sparse Modeling Study Sessions are contributing to the promotion of mutual knowledge exchange among HACARUS members -…

A peek inside our Sparse Modeling Study Sessions

With a combined total of more than 30 full-time employees and interns, our data science team continues to grow, each bringing unique expertise and perspectives. Sparse modeling, which HACARUS core technology is built on, is so diverse that individual differences in knowledge and skills have emerged within our team. In order to complement each other, we have started study sessions to learn more about sparse modeling as part of our everyday business we call these "The Sparse Modeling Study Sessions"…

Evaluation of Human Part Segmentation using the Jetson Nano Developer Kit

In this article, Daisuke Takahashi, a data scientist at HACARUS explains the process of Human Part Segmentation using a high performance edge device; the Jetson Nano Developer Kit.   What is the Jetson Nano Developer Kit? The Jetson Nano is a purpose built edge device by NVIDIA that allows for GPU based operations in a light and cost effective package. Much like a standard computer it allows you to run various applications. For this use case we installed a Linux…