Monday, December 1st 2025

KIOXIA AiSAQ and Memory-Centric AI Innovations Enable AI-Based Automatic Image Recognition for Logistics Processes

Kioxia America, Inc. today announced a collaboration between Kioxia Corporation, Tsubakimoto Chain Co. (Tsubakimoto Chain) and EAGLYS Inc. (EAGLYS) to develop AI-driven image recognition technology that automatically identifies products moving through logistics workflows. This system supports advanced automation and efficiency in logistics, enabling organizations to respond quickly to changing market needs while managing costs and maintaining service quality. KIOXIA AiSAQ 1 and Memory-Centric AI2 technologies are central to this effort, addressing the need for scalable AI adoption as product types continue to expand and diversify. The jointly developed technology will be demonstrated during the 2025 International Robot Exhibition.

With the continued rise of e-commerce transactions, logistics networks are seeing greater volumes and a wider range of products moving through them. At the same time, ongoing labor shortages are driving the need for greater operational efficiency through AI. Traditional image recognition AI systems rely on deep learning models that require parameter tuning and retraining whenever new or seasonal products are introduced. This process is time-consuming and increases both power usage and operational costs, especially when dealing with large product catalogs.
KIOXIA AiSAQ software, combined with KIOXIA's Memory-Centric AI technology, addresses these challenges by storing extensive new product data - including images, labels, and feature information - in high-capacity storage. This allows new product information to be added quickly without retraining the base model. To mitigate longer search times and increased memory requirements as data volume increases, the technology indexes memory-stored data and moves the indexed data into SSD storage, enabling faster and more efficient retrieval in AI systems utilizing Retrieval Augmented Generation (RAG), which is a form of Memory-Centric AI in the field of image recognition.

The technology collaboration between Kioxia, Tsubakimoto Chain, and EAGLYS will be showcased at the 2025 International Robot Exhibition, held December 3-6 at Tokyo Big Sight (Tsubakimoto Chain Booth E6-23). The exhibition is a premier global venue for highlighting the latest automation advances in manufacturing and logistics. At the booth, visitors will see the jointly developed image recognition system in action. As products move along a conveyor, the system captures their image data and rapidly classifies them by referencing stored features and labels. The demonstration shows how logistics sites can manage a wide and continually changing range of products with greater accuracy and efficiency.
Add your own comment

2 Comments on KIOXIA AiSAQ and Memory-Centric AI Innovations Enable AI-Based Automatic Image Recognition for Logistics Processes

#1
_JP_
And standards-recognized (globally) 2-D and 3-D codes needed improvement because...?
Posted on Reply
#2
TheGeekn°72
_JP_And standards-recognized (globally) 2-D and 3-D codes needed improvement because...?
because money
btarunrAt the booth, visitors will see the jointly developed image recognition system in action. As products move along a conveyor, the system captures their image data and rapidly classifies them by referencing stored features and labels. The demonstration shows how logistics sites can manage a wide and continually changing range of products with greater accuracy and efficiency.
Established systems used in logistics throughout the world have already massive throughput capacity, do they really hope that customers whose job is literally to make logistics as efficient and effective as possible will fall for a system that will require an AI, the hardware to run the AI, probably a licence/subscription to run it, potentially prone to errors when, as @_JP_ pointed out, barcodes/QR codes and other international standards are already processed through dumb scanners for a fraction of the price?
memory-centric AI
the fuck even is that???
Posted on Reply
Nov 18th, 2026 23:26 PST change timezone

New Forum Posts

Popular Reviews

TPU on YouTube

Controversial News Posts