Research Article | Published: 30 August 2024

Enhancing bark texture analysis and species classification with deep learning models: Wide residual networks and convNeXt

Rohini Bhusnurmath and Shaila Doddamani

Indian Journal of Forestry | Volume: 47 | Issue: 2 | Page No. 94-103 | 2024
DOI: https://doi.org/10.54207/bsmps1000-2024-1SB03B | Cite this article

Abstract

This study delves into the analysis of bark texture images of common trees using deep learning methods to efficiently classify different wood species. With applications spanning from construction to furniture manufacturing, efficient and precise wood species classification is vital for effective forestry management and the timber trade. The research centers on a dataset featuring images of 50 distinct wood species, each characterized by unique texture patterns. Two deep learning models, Wide Residual Networks (WRN) and ConvNeXt, are employed and compared for their analysis purposes. Results consistently demonstrate WRN's superior performance, attributed to its architectural design and effective training strategy in capturing intricate texture patterns. Notably, WRN achieves impressive efficiency alongside high accuracy, precision, and recall rates of 97.23%, 97.29%, and 97.23%, respectively. WRN's success over the pre-processed dataset underscores its versatility and robustness in handling complex texture patterns. Overall, the study showcases the transformative potential of deep learning in revolutionizing tree species classification.

Keywords

Bark texture classification, BarkVN-50, CNN, Improved Random search algorithm, Mobile Net, Texture classification, VGG16, WRN

Access Options

250/-

Buy Full Access in HTML Format

Instant access to the full article.

Get access to the full version of this article. Buy Full Access in HTML Format

Abstract Keywords References About this Article

Fig. 1

Overall model work flow

Fig. 2

Detailed view of classes in dataset

Fig. 3

BarkVN-50 texture dataset images of each class

Fig. 4

Block diagram of WRN model with ImpRSA

Fig. 5

Accuracy, loss, and precision-recall curves of WRN model

Fig. 6

Accuracy, loss, and precision-recall curves of ConvNext model

Fig. 7

Confusion matrix of WRN model (subset of 10 classes of all 50 classes)

Fig. 8

Confusion matrix of ConvNeXt model (subset of 10 classes of all 50 classes)

References

1. Aggarwal, A. and Kumar, M., 2020. Image surface texture analysis and classification using deep learning. Multimedia Tools and Applications, 80, pp.1289-1309.

Google Scholar

2. Bhusnurmath, R.A. and Doddamani, S., 2023a. Bark Texture Classification Using Deep Transfer Learning. In International Conference on Multi-disciplinary Trends in Artificial Intelligence, pp.407-420. Cham: Springer Nature Switzerland. https://doi.org/10.1007/978-3-031-36402-0_38

Google Scholar

3. Bhusnurmath, R.A. and Doddamani, S., 2023b. Texture Feature Extraction and Classification Using Machine Learning Techniques. In International Conference on Emerging Research in Computing, Information, Communication and Applications, pp.509-520. Singapore: Springer Nature Singapore. https://doi.org/10.1007/978-981-99-7622-5_35

Google Scholar

4. Cui, Z., Li, X., Li, T. and Li, M., 2023. Improvement and Assessment of Convolutional Neural Network for Tree Species Identification Based on Bark Characteristics. Forests, 14(7), 1292. https://doi.org/10.3390/f14071292

Google Scholar

5. Faizal, S., 2022. Automated identification of tree species by bark texture classification using convolutional neural networks. arXiv preprint arXiv:2210.09290. https://doi.org/10.22214/ijraset.2022.46846

Google Scholar

6. Fekri-Ershad, S., 2020. Bark texture classification using improved local ternary patterns and multilayer neural network. Expert Systems with Applications, 158, 113509. https://www.kaggle.com

Google Scholar

7. Hiremath, P.S. and Bhusnurmath, R.A., 2017a. Texture classification using partial differential equation approach and wavelet transform. Pattern Recognition and Image Analysis, 27, pp.473-479. https://doi.org/10.1134/S1054661817030154

Google Scholar

8. Hiremath, P.S. and Bhusnurmath, R.A., 2017b. Industrial applications of colour texture classification based on anisotropic diffusion. In Recent Trends in Image Processing and Pattern Recognition: First International Conference, RTIP2R 2016, Bidar, India, December 16–17, 2016, Revised Selected Papers 1 (pp.293-304). Springer Singapore. https://doi.org/10.1007/978-981-10-4859-3_27

Google Scholar

9. Ido, J. and Saitoh, T., 2019. CNN-based tree species identification from bark image. In Tenth International Conference on Graphics and Image Processing (ICGIP2018) (Vol. 11069, pp.991-996). SPIE. https://doi.org/10.1117/12.2524213

Google Scholar

10. Juola, J., Hovi, A. and Rautiainen, M., 2023. Classification of tree species based on hyperspectral reflectance images of stem bark. European Journal of Remote Sensing, 56(1), 2161420. https://doi.org/10.1080/22797254.2022.2161420

Google Scholar

11. Kapil, R., Marvasti-Zadeh, S. M., Goodsman, D., Ray, N. and Erbilgin, N., 2022. Classification of Bark Beetle-Induced Forest Tree Mortality using Deep Learning. arXiv preprint arXiv:2207.07241. https://doi.org/10.48550/arXiv.2207.07241

Google Scholar

12. Kim, T.K., Hong, J., Ryu, D., Kim, S., Byeon, S.Y., Huh, W. and Kim, H.S., 2022. Identifying and extracting bark key features of 42 tree species using convolutional neural networks and class activation mapping. Scientific Reports, 12(1), p.4772. https://doi.org/10.1038/s41598-022-08571-9

Google Scholar

13. Misra, D., Crispim-Junior, C. and Tougne, L., 2020. Patch-based CNN evaluation for bark classification. In Computer Vision–ECCV 2020 Workshops: Glasgow, UK, August 23–28, 2020, Proceedings, Part VI 16 (pp.197-212). Springer International Publishing. https://doi.org/10.1007/978-3-030-65414-6_15

Google Scholar

14. Mosaffaei, Z. and Jahani, A., 2021. Modeling of ash (Fraxinus excelsior) bark thickness in urban forests using artificial neural network (ANN) and regression models. Modeling Earth Systems and Environment, 7(3), pp.1443-1452. https://doi.org/10.1007/s40808-020-00869-9

Google Scholar

15. Nisa, M., Shah, J.H., Kanwal, S., Raza, M., Khan, M.A., Damaševičius, R. and Blažauskas, T., 2020. Hybrid malware classification method using segmentation-based fractal texture analysis and deep convolution neural network features. Applied Sciences, 10(14), 4966. https://doi.org/10.3390/app10144966

Google Scholar

16. Ratajczak, R., Bertrand, S., Crispim-Junior, C. F. and Tougne, L., 2019. Efficient bark recognition in the wild. In International conference on computer vision theory and applications (VISAPP 2019).

Google Scholar

17. Wan, Y.Y., Du, J.X., Huang, D.S., Chi, Z., Cheung, Y.M., Wang, X.F. and Zhang, G.J., 2004. Bark texture feature extraction based on statistical texture analysis. In Proceedings of 2004 International Symposium on Intelligent Multimedia, Video and Speech Processing, pp. 482-485. https://doi.org/10.1109/ISIMP.2004.1434106

Google Scholar

About this article

How to cite

Bhusnurmath, R. and Doddamani, S., 2024. Enhancing bark texture analysis and species classification with deep learning models: Wide residual networks and convNeXt. Indian Journal of Forestry, 47(2), pp.94-103. https://doi.org/10.54207/bsmps1000-2024-1SB03B

Publication History

Manuscript Received on 03 August 2024

Manuscript Revised on 23 August 2024

Manuscript Accepted on 28 August 2024

Manuscript Published on 30 August 2024

Share this article

Anyone you share the following link with will be able to read this content: