18234913. SYSTEM AND METHOD FOR GENERATING HYPERSPECTRAL ARTIFICIAL VISION FOR MACHINES simplified abstract (Tata Consultancy Services Limited)

SYSTEM AND METHOD FOR GENERATING HYPERSPECTRAL ARTIFICIAL VISION FOR MACHINES

Organization Name

Tata Consultancy Services Limited

Inventor(s)

Shailesh Shankar Deshpande of Pune (IN)

Kran Sharad Owalekar of Thane West (IN)

Apoorva Khanna of New Delhi (IN)

Mahesh Kshirsagar of Mumbai (IN)

Balamuralidhar Purushothaman of Bangalore (IN)

SYSTEM AND METHOD FOR GENERATING HYPERSPECTRAL ARTIFICIAL VISION FOR MACHINES - A simplified explanation of the abstract

This abstract first appeared for US patent application 18234913 titled 'SYSTEM AND METHOD FOR GENERATING HYPERSPECTRAL ARTIFICIAL VISION FOR MACHINES

Simplified Explanation

Embodiments herein provide a method and system for a hyperspectral artificial vision for machines. The system receives a hyperspectral signal of a target material as an input to a neural network model. The system initializes by selecting the number of primitive layers to be used. The system iteratively cycles through all training data (pixels) and updating weights for each unsuccessful material class prediction. Model with two primitives serves as baseline, after which the system adds another primitive layer and repeats the training procedure. The system keeps repeating these processes until obtains convergence. Where the system come to a halt, the system obtains the optimal number of primitives for the given materials. The generated new color pixel is used as a discriminator to aid in locating the target material. The new artificial color is a mixture of weighted chromatic primitives which are optimized for sensitivity/(Spectral Response Functions) SRFs.

• Hyperspectral artificial vision system for machines
• Neural network model with primitive layers
• Iterative training process for material class prediction
• Optimal number of primitives determined for given materials
• New artificial color pixel used for material location
• Mixture of weighted chromatic primitives for color generation

Potential Applications

The technology can be applied in various fields such as:

• Remote sensing
• Agriculture
• Environmental monitoring
• Medical imaging

Problems Solved

• Improved material classification accuracy
• Enhanced hyperspectral imaging capabilities
• Efficient target material detection

Benefits

• Increased precision in material identification
• Enhanced spectral analysis
• Improved machine vision capabilities

Potential Commercial Applications

Optimized hyperspectral artificial vision technology can be utilized in:

• Quality control in manufacturing
• Food safety inspection
• Pharmaceutical research and development

Possible Prior Art

One possible prior art in this field is the use of hyperspectral imaging for material classification and identification. However, the specific method of utilizing neural network models with primitive layers for hyperspectral artificial vision may be a novel approach.

What are the limitations of the current hyperspectral artificial vision systems in terms of scalability and real-time processing capabilities?

The current hyperspectral artificial vision systems may face limitations in terms of scalability when dealing with large datasets or complex materials. Additionally, real-time processing capabilities may be hindered by the computational requirements of the neural network models used in the system.

How does the addition of primitive layers in the neural network model improve the accuracy and efficiency of material classification in hyperspectral artificial vision systems?

The addition of primitive layers in the neural network model allows for a more nuanced and detailed analysis of the hyperspectral signal, leading to improved accuracy in material classification. By iteratively updating weights for each unsuccessful material class prediction, the system can optimize the number of primitives for the given materials, enhancing the efficiency of the classification process.

Original Abstract Submitted

Embodiments herein provide a method and system for a hyperspectral artificial vision for machines. The system receives a hyperspectral signal of a target material as an input to a neural network model. The system initializes by selecting the number of primitive layers to be used. The system iteratively cycles through all training data (pixels) and updating weights for each unsuccessful material class prediction. Model with two primitives serves as baseline, after which the system adds another primitive layer and repeats the training procedure. The system keeps repeating these processes until obtains convergence. Where the system come to a halt, the system obtains the optimal number of primitives for the given materials. The generated new color pixel is used as a discriminator to aid in locating the target material. The new artificial color is a mixture of weighted chromatic primitives which are optimized for sensitivity/(Spectral Response Functions) SRFs.