Neural Network Detection Model Testing with Error Threshold Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Detection models based on neural networks face inefficiencies due to hidden layers that fail to test signal quality, leading to excessive computing power consumption and errors, especially when interference signals are present, resulting in inaccurate detection results and wasted resources.
Innovation Solution
A method is introduced to test detection models by checking error thresholds and re-acquiring data with interference signals, utilizing multiple hidden layers to extract features efficiently, and determining when to stop computation based on probability thresholds to reduce errors and computing power.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple hidden layers are used in the detection model, then the detection accuracy is improved, but the computing power consumption increases
Solution Approach 1:
The patent applies preliminary action by introducing an error threshold testing mechanism before the full computing process. The system tests whether the error of the detection model output exceeds a threshold after processing by hidden layers. If the error is within the threshold, computation stops early; if not, the system re-acquires the signal and continues computing. This preliminary error checking prevents unnecessary computation while maintaining detection accuracy.
2Reliability
If the detection model processes signals with interference, then comprehensive detection is attempted, but large errors exist in the detection result and resources are wasted
Solution Approach 1:
The patent implements feedback by continuously monitoring the error of the detection model output and comparing it against a predefined threshold. The error is calculated based on the difference between the detection result and the actual signal characteristics. When the error exceeds the threshold, the system feedbacks this information and re-acquires the signal for re-processing, thereby correcting detection errors caused by interference signals and avoiding wasted resources on fundamentally flawed detections.
3Productivity
If hidden layers do not test signal quality, then the computing process continues, but excessive computing power is consumed without improving detection quality
Solution Approach 1:
The patent applies preliminary action by introducing an error threshold testing mechanism before the full computing process. The system tests whether the error of the detection model output exceeds a threshold after processing by hidden layers. If the error is within the threshold, computation stops early; if not, the system re-acquires the signal and continues computing. This preliminary error checking prevents unnecessary computation while maintaining detection accuracy.
Data Source
AI summary
A method of testing detection models based on a neural network includes that (a) acquiring to-be-detected data; (b) inputting the to-be-detected data into the primary hidden layer to generate primary output vectors; (c) determining whether an error of the to-be-detected data is larger than a predefined threshold; (d) when the error of the to-be-detected data is larger than the predefined threshold, returning to the step (a); (e) when the error of the to-be-detected data is less than or equal to the predefined threshold, outputting the primary output vectors into a next secondary hidden layer in sequence to generate corresponding secondary output vectors; (f) determining whether a detection result outputs based on the secondary output vectors. An electronic device and a computer readable storage medium are also provided.