Remote Sensing + AI: JiaHe Info Achieves Disruptive Platform Innovation Based on MindSpore, Winning the 2023 CICAS Award

The first China's Innovation Challenge on Artificial Intelligence Application Scene (CICAS) 2023, jointly organized by the Chinese Association for Artificial Intelligence and the Next-Generation Artificial Intelligence Development Research Center of Ministry of Science and Technology, concluded successfully in Beijing recently. After three rounds of intense competition, JiaHe Info was awarded the second prize in the special competition for intelligent remote sensing with its 'MindSpore-powered JiaHe remote sensing AI interpretation platform and SaaS application'.

This challenge aims to accelerate the core technology breakthrough, product development, standard customization, and industrial cultivation of general-purpose AI, with a focus on addressing significant AI application and industrialization issues. It strives to build the No.1 competition brand in AI application scene innovation and scientific and technological achievement industrialization in China, and contributes to empowering the transformation and upgrading of traditional industries and the high-quality development of the digital economy in China.

Remote sensing + AI, the core competitiveness of JiaHe:

As a pioneering industrialization enterprise, JiaHe Info has always been supported by strong technical capabilities in satellite remote sensing. Based on this, JiaHe Info has created several policy-oriented innovative products that satisfy market and industry needs, providing users with remote sensing data operation management, as well as multi-scenario data services.

Leading algorithm capabilities:

With self-developed algorithms and LuoJiaNET as its core, JiaHe Info deeply integrates with the MindSpore AI framework to build private spectrum libraries based on industry-leading hyperspectral image reconstruction and hyperspatial resolution reconstruction technologies. It enhances application accuracies while also reducing data processing time by approximately 50% and overall costs by roughly 30% by utilizing machine learning techniques to extract hyperspectral images from both multispectral and hyperspectral libraries.

One-stop service capabilities:

Putting user needs first, JiaHe Info constructs a one-stop service mode integrating remote sensing data processing, management, and result distribution. It overcomes the technical challenges of traditional remote sensing image processing, which is time-consuming and labor-intensive, and provides users with a target-oriented remote sensing data management platform that supports one-click management, multi-dimensional window expansion, and automatic running, achieving a convenient, open, flexible, and stable disruptive innovation.

Data standardization capabilities:

Having been deeply involved in the remote sensing field for years, JiaHe Info has established a massive sample library of remote sensing images, with over 40 types of AI algorithms for clutter identification and over 200 application-level algorithms in vertical scenarios integrated. This enables automatic computation of various clutter monitoring types, such as land plot boundaries, crop types, and ecological environment classification. All data products comply with the OGC standard.

Quick delivery capabilities:

JiaHe Info has encapsulated satellite remote sensing big data into standardized APIs, and provides technical support for data APIs and SDK customization services for customers. The entire process of data production, UI design, and visualization platform construction has been standardized, which can reduce the customer labor cost by nearly 80% and enable rapid delivery within hours. The recognition from the Chinese Association for Artificial Intelligence and the Ministry of Science and Technology for the 'MindSpore-powered JiaHe remote sensing AI interpretation platform and SaaS application' will motivate MindSpore to persist in collaborating with JiaHe Info to continuously to enhance technologies, refine scenes, and implement more benchmark cases for the agriculture industry, farmers, and rural areas.