1. Introduction

• To achieve the classification of aero-engines, this paper employs an infrared spectroscopy detection method to measure the spectra of aero-engines’ hot jets, which are significant sources of aero-engines’ infrared radiation, using an FTIR spectrometer. The FT-IR spectrometer’s infrared spectrum offers characteristic molecular-level information about substances, thereby enhancing the scientific of classifying aero-engines based on their infrared spectra.
• This paper presents a SCNN for classifying the hot jet spectrum of aero-engines using a data matching method. The network is based on 1DCNN, and feature similarity is calculated using the Euclidean distance metric. Subsequently, a spectral comparison method is employed for the purpose of performing spectrum classification.
• The objective of this paper is to propose an algorithm for identifying peaks that will optimize the training and prediction speed of the SCNN. The algorithm identifies the peak value in the mid-infrared spectrum data and quantifies the high-frequency peaks, which are subsequently employed as input for the SCNN.

2. Experimental Design and Dataset Production for Hot Jet Infrared Spectrum Measurement of Aero-Engines

3. Architectural Design of Peak Finding Siamese Convolutional Neural Network

4. Experiment and Result

4.1. Performance Measures and Experiment Results

Based on the evaluation criteria for data classification tasks, the performance measures of PF-SCNN for aero-engine hot jet includes Accuracy, Precision, Recall, F1-score and Confusion matrix. An instance is classified as a positive class and predicted to be positive, resulting in a True Positive (TP); if it is classified as negative but predicted as positive, it becomes a False Negative (FN); conversely, if the instance is a negative class and predicted to be positive, it results in a False Positive (FP), while predicting negative correctly leads to True Negative (TN). Based on these assumptions, evaluation criteria such as accuracy rate, recall rate, F1 value and confusion matrix are defined:

Accuracy: The accuracy is the ratio of correctly classified samples to the total number of samples.

$$Accuracy={\displaystyle \frac{TP+TN}{TP+TN+FP+FN}}$$

(14)

Precision: The precision is the ratio of correctly predicted positive samples to all predicted positive samples.

$$Precision={\displaystyle \frac{TP}{TP+FP}}$$

(15)

Recall: The recall is the ratio of the number of samples correctly predicted to be positive to the number of samples that are actually positive.

$$Recall={\displaystyle \frac{TP}{TP+FN}}$$

(16)

F1-score: The F1-score is a composite measure of precision and recall, which considers both aspects to evaluate the overall performance of the model.

$$F1-score={\displaystyle \frac{2^{\ast }{P}^{\ast }R}{P+R}}$$

(17)

where the $P$ stands for $Precision$,while the $R$ stands for $Recall$.

Confusion Matrix: The confusion matrix presents the classification results of various categories by the classifier, encompassing TP, FP, TN and FN. It visually demonstrates the disparity between actual and predicted values, with the diagonal elements indicating the number of accurate predictions for each category made by the classifier. Table 4 offers a detailed breakdown of each component in the confusion matrix.

To validate the algorithm’s effectiveness, training and prediction experiments were conducted on a dataset comprising six types of aero-engine spectra. The experiments took place on a Windows 10 workstation equipped with 32 GB of RAM, an Intel Core i7-8750H processor, and a GeForce RTX 2070 graphics card.

Table 5 provides detailed parameters for PF-SCNN:

According to the Table 5 parameters, we conducted training and label prediction on the data set, and obtained experimental results as shown in Table 6:

Figure 5. PF-SCNN spectrum matching classification experiment results, where the left figure is the loss function transformation curve, the right figure is the correct rate transformation curve, the blue line is the training data, orange is the verification data.

Figure 5. PF-SCNN spectrum matching classification experiment results, where the left figure is the loss function transformation curve, the right figure is the correct rate transformation curve, the blue line is the training data, orange is the verification data.

The PF-SCNN, as designed in this paper, effectively classifies the six types of aero-engine hot jet spectrum dataset with 99.46% accuracy. The model demonstrates high precision (99.77%) and recall (99.56%), accurately predicting both positive classes and actual positive samples. The confusion matrix provides insights into prediction performance for each class. Analysis of the F1 score (99.66%) shows a strong balance between accuracy and recall, while Loss and accuracy converge rapidly within 500 training sessions—taking 2757.8s for training and 71s for label prediction per data.

Despite encountering special cases such as aero-engine failure during experiment, our PF-SCNN demonstrates robustness with minimal impact on overall classification accuracy when handling error data within spectrum data set.

4.3. Ablation Experiment Analysis

(1) Peak Feature Effectiveness: Integrating peak features with traditional classifiers validates their impact on algorithm enhancement for classification. We identified peaks in our data set and obtained experimental results, as depicted in Figure 7:

In Figure (a), the red points represent the peak positions identified by the peak finding algorithm; in (b), the blue histogram illustrates the distribution of these peak positions within the data set; and in (c), the red line indicates the locations of points with higher frequency in the spectrum data set. Following a comprehensive statistical analysis of these peak positions, a total of 56 peaks were identified. Subsequently, data corresponding to these 56 peak positions from each dataset were computed, and a classification experiment was conducted using SVM and XGBoost classifier methods as part of our previous study, resulting in detailed classification results presented in Table 10:

Based on the experimental results, the extracted peak data demonstrates significant efficacy. When compared with CO₂ feature vector classifiers, all indices exhibit notable enhancements. The experimental findings suggest that feature extraction using the peak finding algorithm is highly effective for classification tasks.

(2) SCNN model

We use the same parameters of the PF-SCNN model to input data in the mid-infrared region into SCNN for training and prediction, and obtain Table 11:

Figure 8. SCNN spectrum matching classification experiment results, where the left figure is the loss function transformation curve, the right figure is the correct rate transformation curve, the blue line is the training data, the orange line is the verification data.

Figure 8. SCNN spectrum matching classification experiment results, where the left figure is the loss function transformation curve, the right figure is the correct rate transformation curve, the blue line is the training data, the orange line is the verification data.

When the entire mid-infrared spectrum was used as input, the training time amounted to 19103.60 seconds. It is evident that the SCNN model can achieve commendable accuracy in both training and prediction. However, owing to data quantity issues, the SCNN model demands substantial computing resources and time.

(3) Optimizer selection:

In deep learning network training, we input commonly used optimizers such as RMSProp, Adam, Nadam, SGD, Adagrad, and Adadelta into the base network model for comparison on the spectrum data used in this paper to determine the most suitable optimizer. The optimizer parameters are shown in Table 12:

We conducted 200 training tests on the peak data of the spectrum dataset using various optimizers on the base network, resulting in Table 13:

Figure 9. Network training and validation of Loss function and Accuarcy variation curves on spectrum data set by different optimizers: RMSProp in blue, Adam in orange, Nadam in green, SGD in red, Adagrad in purple, and Adadelta in brown.

Figure 9. Network training and validation of Loss function and Accuarcy variation curves on spectrum data set by different optimizers: RMSProp in blue, Adam in orange, Nadam in green, SGD in red, Adagrad in purple, and Adadelta in brown.

As depicted in the graph, both RMSProp and Adam exhibit rapid convergence on the loss function curve and accuracy curve after 200 iterations of model training, leading to high prediction accuracy.

(4) Learning rate selection:

The choice of learning rate is a crucial step in training a neural network. In our experiments, we tried different learning rate values and observed their impact on the model’s performance. Starting from 0.001, we gradually increased the learning rate using a power of 10. In this way, we obtained a series of results and summarized them in Table 14 for detailed comparison and analysis. This helps us find the optimal learning rate value that best fits the data set and model architecture, thereby improving training efficiency and model performance:

Figure 10. Network training and validation of RMSProp with different learning rates on spectrum data sets loss function and accuarcy change curve: Among them, blue represents the learning rate of 0.001, orange represents the learning rate of 0.0001, green represents the learning rate of 0.00001, and red represents the learning rate of 0.000001.

Figure 10. Network training and validation of RMSProp with different learning rates on spectrum data sets loss function and accuarcy change curve: Among them, blue represents the learning rate of 0.001, orange represents the learning rate of 0.0001, green represents the learning rate of 0.00001, and red represents the learning rate of 0.000001.

Based on the variations in the loss function and accuracy curves for a given optimizer under different learning rates, as well as considering the final accuracy value, it is apparent that RMSProp demonstrates optimal performance on the spectrum data set when utilizing a learning rate of 0.0001. It is consistent with the learning rate we have adopted.

(5) Running time:

Deep learning methods necessitate longer model training times due to their typical utilization of large datasets and complex network structures. In contrast to traditional classifier methods, deep learning algorithms require more iterations for parameter adjustment and model optimization in order to achieve higher accuracy and generalization capability. In practical applications, we conducted a detailed comparison of prediction times for different methods on datasets and compiled the results into a test time comparison table Table 15. These data clearly illustrate the differences in prediction times required by various algorithms when processing the same dataset, providing valuable references for further analysis and evaluation:

The data in Table 15 indicates that CO₂ feature vector classifiers exhibit fast running times. However, the PF-SCNN does not offer significant advantages in terms of running time, as it requires matching prediction data with trained data during the network’s prediction stage, resulting in substantial consumption of computing time and memory. Although introducing peak values significantly enhances algorithmic speed compared to using only the SCNN model for prediction, the one-to-one matching method still demands considerable computing time. Based on these findings regarding running times, our future research will focus on extracting stable and distinct features from each type of aero engine’s hot jet spectrum to reduce predicted data volume and prediction time.

5. Summary

References

1. ROZENSTEIN O, PUCKRIN E, ADAMOWSKI J. Development of a new approach based on midwave infrared spectroscopy for post-consumer black plastic waste sorting in the recycling industry. Waste Management, 2017, 68: 38-44. [CrossRef]
2. HOU X, LV S, CHEN Z, et al. Applications of Fourier transform infrared spectroscopy technologies on asphalt materials. Measurement, 2018: 304-316. [CrossRef]
3. OZAKI Y. Infrared Spectroscopy—Mid-infrared, Near-infrared, and Far-infrared/Terahertz Spectroscopy. Analytical Sciences,2021: 1193-1212. [CrossRef]
4. Jang, H.-D.; Kwon, S.; Nam, H.; Chang, D.E. Semi-Supervised Autoencoder for Chemical Gas Classification with FTIR Spectrum. Sensors,2024,24, 3601. [CrossRef]
5. UDDIN Md P, MAMUN Md A, HOSSAIN Md A. PCA-based Feature Reduction for Hyperspectral Remote Sensing Image Classification. IETE Technical Review, 2021: 377-396.
6. XIA J, BOMBRUN L, ADALI T, et al. Spectral–Spatial Classification of Hyperspectral Images Using ICA and Edge-Preserving Filter via an Ensemble Strategy. IEEE Transactions on Geoscience and Remote Sensing, 2016: 4971-4982. [CrossRef]
7. JIA S, ZHAO Q, ZHUANG J, et al. Flexible Gabor-Based Superpixel-Level Unsupervised LDA for Hyperspectral Image Classification. IEEE Transactions on Geoscience and Remote Sensing, 2021: 10394-10409. [CrossRef]
8. Zhang Y, Li T. Three different SVM classification models in Tea Oil FTIR Application Research in Adulteration Detection//Journal of Physics: Conference Series. IOP Publishing, 2021, 1748(2): 022037.
9. Chen T, Guestrin C. XGBoost: A scalable tree boosting system//Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining. 2016: 785-794.
10. Breiman L. Random Forests. Machine learning, 2001, 45: 5-32.
11. Kumaravel ArulRaj,Muthu Karthikeyan,Deenadayalan Narmatha．A View of Artificial Neural Network Models in Different Application Areas．E3S Web of Conferences,2021,287(a).
12. LI X, LI Z, QIU H, et al. An overview of hyperspectral image feature extraction, classification methods and the methods based on small samples. Applied Spectroscopy Reviews, 2021: 1-34. [CrossRef]
13. CHEN Y, LIN Z, ZHAO X, et al. Deep Learning-Based Classification of Hyperspectral Data. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2014: 2094-2107.
14. ZHOU W, KAMATA S I, WANG H, et al. Multiscanning-Based RNN-Transformer for Hyperspectral Image Classification. IEEE Transactions on Geoscience and Remote Sensing, 2023: 1-19. [CrossRef]
15. Hu, H.; Xu, Z.; Wei, Y.; Wang, T.; Zhao, Y.; Xu, H.; Mao, X.; Huang, L. The Identification of Fritillaria Species Using Hyperspectral Imaging with Enhanced One-Dimensional Convolutional Neural Networks via Attention Mechanism. Foods 2023, 12, 4153. [CrossRef]
16. Ma, Y.; Lan, Y.; Xie, Y.; Yu, L.; Chen, C.; Wu, Y.; Dai, X. A Spatial–Spectral Transformer for Hyperspectral Image Classification Based on Global Dependencies of Multi-Scale Features. Remote Sens. 2024, 16, 404. [CrossRef]
17. JIA S, JIANG S, LIN Z, et al. A Semisupervised Siamese network for Hyperspectral Image Classification. IEEE Transactions on Geoscience and Remote Sensing, 2022: 1-17. [CrossRef]
18. ONDRASOVIC M, TARABEK P. Siamese Visual Object Tracking: A Survey. IEEE Access, 2021: 110149-110172. [CrossRef]
19. XU T, FENG Z, WU X J, et al. Toward Robust Visual Object Tracking With Independent Target-Agnostic Detection and Effective Siamese Cross-Task Interaction. vol. 32, pp. 1541-1554, 2023. [CrossRef]
20. WANG L, WANG L, WANG Q, et al. SSA-SiamNet: Spectral–Spatial-Wise Attention-Based Siamese network for Hyperspectral Image Change Detection. IEEE Transactions on Geoscience and Remote Sensing, 2022: 1-18.
21. MELEKHOV I, KANNALA J, RAHTU E. Siamese network features for image matching//2016 23rd International Conference on Pattern Recognition (ICPR), Cancun. 2016.
22. LI Y, CHEN C L P, ZHANG T. A Survey on Siamese network: Methodologies, Applications, and Opportunities. IEEE Transactions on Artificial Intelligence, 2022: 994-1014.
23. HUANG L, CHEN Y. Dual-Path Siamese CNN for Hyperspectral Image Classification With Limited Training Samples. IEEE Geoscience and Remote Sensing Letters, 2021: 518-522. [CrossRef]
24. MIAO J, WANG B, WU X, et al. Deep Feature Extraction Based on Siamese network and Auto-Encoder for Hyperspectral Image Classification. IEEE Conference Proceedings,IEEE Conference Proceedings, 2019.
25. NANNI L, MINCHIO G, BRAHNAM S, et al. Experiments of Image Classification Using Dissimilarity Spaces Built with Siamese networks. Sensors, 2021, 21(5): 1573. [CrossRef]
26. LIU B, YU X, ZHANG P, et al. Supervised Deep Feature Extraction for Hyperspectral Image Classification. IEEE Transactions on Geoscience and Remote Sensing, 2018: 1909-1921. [CrossRef]
27. LI X, LI Z, QIU H, et al. An overview of hyperspectral image feature extraction, classification methods and the methods based on small samples. Applied Spectroscopy Reviews, 2021: 1-34. [CrossRef]
28. KRUSE FredA, KIEREIN-YOUNG K S, BOARDMAN JosephW. Mineral mapping at Cuprite, Nevada with a 63-channel imaging spectrometer. Photogrammetric Engineering and Remote Sensing,Photogrammetric Engineering and Remote Sensing, 1990.
29. BROMLEY J, BENTZ J W, BOTTOU L, et al. SIGNATURE VERIFICATION USING A “SIAMESE” TIME DELAY NEURAL NETWORK//Series in Machine Perception and Artificial Intelligence,Advances in Pattern Recognition Systems Using Neural Network Technologies. 1994: 25-44.
30. THENKABAIL PrasadS, KRISHNA M, TURRAL H. Spectral Matching Techniques to Determine Historical Land-use/Land-cover (LULC) and Irrigated Areas Using Time-series 0.1-degree AVHRR Pathfinder Datasets. Photogrammetric Engineering and Remote Sensing,Photogrammetric Engineering and Remote Sensing, 2007.
31. SOHN K. Improved deep metric learning with multi-class N-pair loss objective. Neural Information Processing Systems,Neural Information Processing Systems, 2016.
32. Homan D C, Cohen M H, Hovatta T, et al. MOJAVE. XIX. Brightness Temperatures and Intrinsic Properties of Blazar Jets. The Astrophysical Journal, 2021, 923(1): 67. [CrossRef]
33. ZHOU W, ZHANG J, JIE D. The research of near infrared spectral peak detection methods in big data era//2016 ASABE International Meeting. 2016.
34. Tieleman, T., & Hinton, G. (2012). Lecture 6.5-rmsprop: divide the gradient by a running average of its recent magnitude. COURSERA: Neural networks for machine learning, 4(2), 26–31.