Mainstream Smoke Spectral Data Sensory Evaluation Method
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current sensory evaluation methods for mainstream smoke spectral data are laborious, inefficient, and unstable due to reliance on human expertise and external factors, making them inaccurate and cumbersome.
Innovation Solution
A method that enhances spectral data through processing techniques such as data augmentation, outlier elimination, denoising, and deep learning using convolutional neural networks to extract shallow and deep sensory quality results, improving accuracy and stability by constructing classification models based on PCA and SVM, and utilizing deep residual CNNs for spatial characteristic extraction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If expert evaluation is used for sensory evaluation of mainstream smoke spectral data, then human expertise can be utilized, but the evaluation becomes laborious, inefficient, and unstable
Solution Approach 1:
The patent replaces the mechanical system of human expert evaluation with an automated computer-based spectral analysis system. The system uses spectral data processing, characteristic extraction, and classification models to automatically evaluate mainstream smoke quality, eliminating the need for manual expert assessment while maintaining or improving accuracy and significantly increasing efficiency.
Solution Approach 2:
The evaluation system performs self-assessment by automatically processing spectral data, extracting characteristics, and generating evaluation results without requiring human intervention. The system uses built-in algorithms and models to independently complete the entire evaluation process, making it self-sufficient and eliminating the labor-intensive nature of expert evaluation.
2Reliability
If expert evaluation is used for sensory evaluation of mainstream smoke spectral data, then human judgment can be applied, but the evaluation is affected by expert factors and external environment
Solution Approach 1:
The patent replaces the unreliable human expert evaluation system with a stable automated computational system. The system uses consistent algorithms, mathematical models, and processing procedures that are not influenced by human factors or external environmental conditions, thereby significantly improving evaluation reliability and stability.
Solution Approach 2:
The patent transforms the evaluation process from subjective human judgment to objective parameter-based analysis. By converting sensory evaluation into quantitative spectral parameter analysis with defined thresholds and classification criteria, the system eliminates variability introduced by human factors while maintaining evaluation comprehensiveness.
3Ease of manufacture
If traditional spectral analysis is used, then simple processing can be applied, but it cannot effectively quantify the correlation between smoke composition and sensory evaluation
Solution Approach 1:
The patent segments the spectral analysis process into multiple distinct stages: data preprocessing, characteristic extraction (both shallow spectral characteristics and deep spatial characteristics), classification modeling, and evaluation result generation. This segmented approach allows each stage to be optimized independently, achieving both processing efficiency and high correlation quantification accuracy.
Solution Approach 2:
The patent extends the analysis from traditional one-dimensional spectral data to multi-dimensional analysis by extracting both shallow spectral characteristics and deep spatial characteristics. This dimensional expansion enables the system to capture more comprehensive information about smoke composition and its relationship with sensory evaluation, significantly improving correlation quantification accuracy.
Data Source
Figure 1
AI summary
The present invention provides a sensory evaluation method for spectral data of mainstream smoke. The method includes: performing a data enhancement on spectral data of mainstream smoke of a plurality of cigarettes; extracting a shallow spectral characteristic from the spectral data of the mainstream smoke of each cigarette; obtaining a shallow sensory quality result of the spectral data of the mainstream smoke of each cigarette based on the spectral data of the mainstream smoke of each cigarette and the shallow spectral characteristic; extracting deep spatial characteristics from the spectral data of the mainstream smoke of each cigarette; obtaining a deep sensory quality result based on the spectral data of the mainstream smoke of each cigarette and the deep spatial characteristics; obtaining a comprehensive sensory quality result according to the shallow sensory quality result and the deep sensory quality result. In the present invention, the sensory evaluation method for the spectral data of the mainstream smoke respectively extracts spectral and spatial characteristics from the shallower to the deeper, and automatically and directly obtains the sensory evaluation results of the mainstream smoke by a fused spectrum-spatial classification framework, so as to achieve accurate screening of unknowns in the mainstream smoke.