18676409. INFERENCE DEVICE simplified abstract (FUJIFILM Corporation)
Contents
INFERENCE DEVICE
Organization Name
Inventor(s)
Seiji Tanaka of Saitama-shi (JP)
INFERENCE DEVICE - A simplified explanation of the abstract
This abstract first appeared for US patent application 18676409 titled 'INFERENCE DEVICE
The abstract describes an inference device with two arithmetic modules that perform arithmetic processing, including convolution and pooling processes. The first module processes first row data items generated by dividing first image data, while the second module processes second row data items generated by dividing second image data. The first and second image data have different numbers of channels.
- The device includes a first arithmetic module and a second arithmetic module for executing arithmetic processing.
- The first arithmetic module has a first memory storing first row data items and first arithmetic units for executing a first convolution process.
- The second arithmetic module has a second memory storing second row data items and second arithmetic units for executing a second convolution process.
- The first and second image data have varying numbers of channels.
- The device performs convolution and pooling processes on the divided row data items.
Potential Applications: - Image processing - Pattern recognition - Machine learning algorithms
Problems Solved: - Efficient processing of image data with varying channel numbers - Enhanced convolution and pooling processes
Benefits: - Improved accuracy in image recognition tasks - Faster processing of large datasets - Versatile application in various industries
Commercial Applications: Title: Advanced Image Processing Inference Device for Machine Learning Applications This technology can be used in: - Autonomous vehicles - Medical imaging - Surveillance systems - Robotics
Questions about the technology: 1. How does the device handle image data with different numbers of channels? 2. What are the advantages of using convolution and pooling processes in image processing?
Original Abstract Submitted
An inference device includes a first arithmetic module and a second arithmetic module that execute arithmetic processing including a convolution process and a pooling process. The first arithmetic module includes a first memory that stores a plurality of first row data items generated by dividing first image data for each first number of pixels in a row direction and a plurality of first arithmetic units that execute a first convolution process on the plurality of first row data items. The second arithmetic module includes a second memory that stores a plurality of second row data items generated by dividing second image data for each second number of pixels in the row direction and a plurality of second arithmetic units that execute a second convolution process on the plurality of second row data items. The first image data and the second image data have different numbers of channels.
