AI Electronic Device Filter Rotation Scaling Texture Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current AI systems, particularly those using deep neural networks, face challenges in recognizing textures due to variations in rotation, size, and magnification, requiring large datasets and high-performance hardware, which can lead to decreased recognition rates and increased learning times.
Innovation Solution
An electronic device employs a modified filter approach by rotating and scaling elements within a convolutional neural network (CNN) model to identify input images, generating multiple filters to improve texture recognition accuracy without the need for extensive data or high computational resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a deep neural network is used to recognize textures with variations in rotation, size, and magnification, then recognition accuracy may be improved, but the amount of data required increases and learning time is lengthened
Solution Approach 1:
The patent applies preliminary action by pre-processing the input image to normalize rotation, size, and magnification variations before texture recognition. This preparation work is done in advance, so the deep neural network doesn't need to learn all possible variations, reducing learning time while maintaining accuracy
Solution Approach 2:
The patent changes parameters by transforming the input image to a standard orientation, size, and magnification before recognition. This parameter normalization allows the model to work with consistent inputs, reducing the need for extensive training data and learning time
2Measurement precision
If the amount of training data is increased to improve recognition rate across different rotations and sizes, then recognition accuracy is improved, but high-performance hardware is required and processing cost increases
Solution Approach 1:
The system performs preliminary normalization of input images to standard rotation, size, and magnification before recognition. This pre-processing reduces the variability that would otherwise require extensive training data and computational power to handle
Solution Approach 2:
By changing the parameters of the input image (rotation angle, size, magnification) to standardized values before recognition, the system reduces the computational complexity and data requirements, achieving good recognition rates without requiring high-performance hardware
3Measurement precision
If a deep neural network processes large amounts of data to achieve high recognition rate, then accuracy is improved, but device complexity and hardware requirements increase
Solution Approach 1:
The patent applies preliminary action by pre-processing the input image to normalize rotation, size, and magnification variations before texture recognition. This preparation work is done in advance, so the deep neural network doesn't need to learn all possible variations, reducing learning time while maintaining accuracy
Solution Approach 2:
The patent changes parameters by transforming the input image to a standard orientation, size, and magnification before recognition. This parameter normalization allows the model to work with consistent inputs, reducing the need for extensive training data and learning time
Data Source
AI summary
The present disclosure relates to an artificial intelligence (AI) system that utilizes a machine learning algorithm and an application thereof. Disclosed is an electronic device. The electronic device comprises: a memory in which a first filter for identifying an input image is stored; and a processor for rotating between a plurality of elements included in the memory in which the first filter is stored and a plurality of elements included in the first filter, obtaining at least one second filter by scaling a filter region including at least some of the plurality of elements, and identifying the input image on the basis of a result value obtained by performing convolution on a pixel value included in the input image with each of the first filter and the second filter.