Adaptive Neural Network for Display Picture Quality
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing picture quality improving neural network models require subjective human classification, making them inadequate for recent technological developments and difficult to adapt to changes in learning image databases.
Innovation Solution
A display apparatus and control method that uses a prediction neural network model to classify input images into clusters based on loss information, allowing for the creation of an adaptive neural network model with improved picture quality without human intervention, by training neural network models corresponding to each cluster and applying weight values based on probability values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If subjective classification work by human is used to train neural network models, then picture quality improvement can be achieved, but the system becomes difficult to adapt to recent technological developments and requires repeated classification work when learning image databases change
Solution Approach 1:
The system enables neural network models to automatically classify images into clusters based on loss information without requiring human subjective classification. The models self-organize and adapt to different image databases through automated clustering, eliminating the need for repeated human classification work when databases change
Solution Approach 2:
The system changes the classification parameter from subjective human assessment to objective loss information-based clustering. By using loss values as the classification criterion, the system achieves both accurate picture quality assessment and automatic adaptability to different image databases
2Measurement precision
If multiple neural network models are trained for different image clusters, then picture quality improvement accuracy is enhanced, but the complexity of the system increases
Solution Approach 1:
The system segments the image database into multiple clusters based on loss information, with each cluster trained by a dedicated neural network model. This segmentation approach improves classification accuracy for different image types while maintaining manageable system complexity through automated clustering
Solution Approach 2:
The system dynamically determines the number of clusters and assigns images to appropriate clusters based on loss information. This dynamic approach allows the system to adapt to different image databases without fixed structural constraints, balancing accuracy improvement with system complexity
3Adaptability or versatility
If loss information is used for automated image classification, then adaptability to image database changes is improved, but the computational complexity increases
Solution Approach 1:
The system performs preliminary clustering of images into groups based on loss information before applying specific neural network models. This preliminary organization reduces the computational burden during actual image processing while maintaining high adaptability to different databases
Data Source
AI summary
A display apparatus, includes: a memory configured to store at least one instruction; and one or more processors configured to execute the at least one instruction to cause the display apparatus to: obtain weight value information for clusters classified according to picture quality by inputting an input image in a prediction neural network model; obtain an adaptive neural network model by respectively applying the weight value information to neural network models corresponding to the clusters; and obtain an output image with improved picture quality by inputting the input image in the adaptive neural network model, wherein the prediction neural network model is a model trained to output probability values for the clusters based on loss information for output images obtained by inputting learning images into the neural network models.