Server Compression Rate Control for AI Accuracy Stability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The challenge is to increase the compression rate of moving image data while minimizing the deterioration of AI analysis accuracy during transmission, as existing methods often result in prolonged periods of accuracy degradation due to over-compression.
Innovation Solution
A server system that detects signs of accuracy deterioration by analyzing accuracy information and adjusts the compression rate by setting upper and lower bounds for quantization values, allowing for dynamic control of compression to maintain high accuracy and efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the compression rate is increased to improve transmission efficiency, then data transmission efficiency is improved, but analysis accuracy deteriorates
Solution Approach 1:
The patent implements dynamic adjustment of compression rates by setting upper and lower bounds that are changed based on stability of analysis accuracy. The system transitions from static compression to dynamic control, adapting the compression rate to maintain accuracy while improving transmission efficiency.
Solution Approach 2:
The patent uses feedback from accuracy information to control the compression rate. By detecting signs of deterioration in analysis accuracy and using this feedback to adjust compression bounds, the system creates a closed-loop control mechanism that resolves the contradiction between compression efficiency and accuracy maintenance.
2Quantity of substance
If the compression rate is increased to reduce data size, then data size is reduced, but analysis accuracy deteriorates
Solution Approach 1:
The system dynamically adjusts compression rates through upper and lower bound changes based on accuracy stability detection, enabling the data size to be reduced without permanently sacrificing analysis accuracy.
Solution Approach 2:
The patent changes the compression rate parameter dynamically by adjusting upper and lower bounds based on detected accuracy deterioration signs, allowing optimization of both data size and analysis accuracy through parameter adaptation.
3Productivity
If the compression rate is increased to improve processing efficiency, then processing efficiency is improved, but analysis accuracy deteriorates
Solution Approach 1:
The patent implements dynamic control of compression rates with upper and lower bounds that adapt based on accuracy stability, enabling processing efficiency improvements while preventing analysis accuracy deterioration through real-time adjustment.
Solution Approach 2:
The system uses feedback from accuracy detection to control compression rate adjustments, creating a feedback mechanism that resolves the contradiction between processing efficiency and analysis accuracy by adapting compression based on observed performance.
Data Source
AI summary
A server includes: a memory; and a processor coupled to the memory and configures to: receive a frame image of a moving image compressed by an information processing device; detect a sign of deterioration in analysis accuracy based on accuracy information that represents certainty of an analysis result by artificial intelligence for the frame image; transmit, to the information processing device, instruction information of controlling a compression rate in compression of the moving image by the information processing device between an upper bound and a lower bound; and change the upper bound and the lower bound according to stability of the analysis accuracy based on the detection of a sign of deterioration in the analysis accuracy.