In practice, the identification of cashmere and wool fibers is still manual. With the expertise of experts, the shape of the fiber surface is observed and distinguished. This manual identification method is difficult because it requires at least years of training for qualified inspectors to observe a large number of fiber images, which is time-consuming and cannot avoid subjective human intervention. Although researchers have proposed many alternative identification methods, none of them can combine efficiency and cost. Therefore, it is particularly important to find a convenient and efficient identification method. The main purpose of this paper is to fuse the decision tree algorithm to study the identification system of fine wool and cashmere. This paper systematically introduces the method of cashmere and wool fiber identification based on image processing technology and feature extraction technology through fiber apparent morphological features. The process of fiber recognition is improved, and the features used for fiber recognition are obtained. The relevant theories and implementation methods are introduced in detail. Two key steps, image processing and feature extraction, are realized based on the theory. Finally, the effectiveness and feasibility of the method are verified through experimental tests, and good results are achieved.

Yaning Yang. A Fine Wool and Cashmere Identification System Incorporating Decision Tree Algorithms. Machine Learning Theory and Practice (2022), Vol. 3, Issue 2: 12-23. https://doi.org/10.38007/ML.2022.030202.

[1] Thai Doan Chuong, V. Jeyakumar, Guoyin Li, Daniel Woolnough: Exact SDP reformulations of adjustable robust linear programs with box uncertainties under separable quadratic decision rules via SOS representations of non-negativity. J. Glob. Optim. 81(4): 1095-1117 (2021). https://doi.org/10.1007/s10898-021-01050-x

[2] Beverly P. Woolf: Introduction to IJAIED Special Issue, FATE in AIED. Int. J. Artif. Intell. Educ. 32(3): 501-503 (2022). https://doi.org/10.1007/s40593-022-00299-x

[3] Kazim Yildiz: Identification of wool and mohair fibres with texture feature extraction and deep learning. IET Image Process. 14(2): 348-353 (2020). https://doi.org/10.1049/iet-ipr.2019.0907

[4] Stefano V. Albrecht, Michael J. Wooldridge: Multi-agent systems research in the United Kingdom. AI Commun. 35(4): 269-270 (2022). https://doi.org/10.3233/AIC-229003

[5] Anna Gautier, Michael J. Wooldridge: Understanding Mechanism Design - Part 3 of 3: Mechanism Design in the Real World: The VCG Mechanism. IEEE Intell. Syst. 37(1): 108-109 (2022). https://doi.org/10.1109/MIS.2021.3129085

[6] Liron David, Avishai Wool: Rank estimation with bounded error via exponential sampling. J. Cryptogr. Eng. 12(2): 151-168 (2022). https://doi.org/10.1007/s13389-021-00269-4

[7] Vess L. Johnson, Richard W. Woolridge, Angelina I. T. Kiser, Katia Guerra: The Impact of Coproduction Resentment on Continuation Intention. J. Comput. Inf. Syst. 62(2): 410-421 (2022). https://doi.org/10.1080/08874417.2021.1971578

[8] Lydia Barnes, Erin Goddard, Alexandra Woolgar: Neural Coding of Visual Objects Rapidly Reconfigures to Reflect Subtrial Shifts in Attentional Focus. J. Cogn. Neurosci. 34(5): 806-822 (2022). https://doi.org/10.1162/jocn_a_01832

[9] Erin Goddard, Thomas A. Carlson, Alexandra Woolgar: Spatial and Feature-selective Attention Have Distinct, Interacting Effects on Population-level Tuning. J. Cogn. Neurosci. 34(2): 290-312 (2022). https://doi.org/10.1162/jocn_a_01796

[10] Amanda K. Robinson, Anina N. Rich, Alexandra Woolgar: Linking the Brain with Behavior: The Neural Dynamics of Success and Failure in Goal-directed Behavior. J. Cogn. Neurosci. 34(4): 639-654 (2022). https://doi.org/10.1162/jocn_a_01818

[11] Daniel Woolnough, Niroshan Jeyakumar, Guoyin Li, Clement T. Loy, Vaithilingam Jeyakumar: Robust Optimization and Data Classification for Characterization of Huntington Disease Onset via Duality Methods. J. Optim. Theory Appl. 193(1): 649-675 (2022). https://doi.org/10.1007/s10957-021-01835-w

[12] Ayla Stein Kenfield, Liz Woolcott, Santi Thompson, Elizabeth Joan Kelly, Ali Shiri, Caroline Muglia, Kinza Masood, Joyce Chapman, Derrick Jefferson, Myrna E. Morales: Toward a definition of digital object reuse. Digit. Libr. Perspect. 38(3): 378-394 (2022). https://doi.org/10.1108/DLP-06-2021-0044

[13] James Gale, Max Seiden, Deepanshu Utkarsh, Jason Frantz, Rob Woollen, Çagatay Demiralp: Sigma Workbook: A Spreadsheet for Cloud Data Warehouses. Proc. VLDB Endow. 15(12): 3670-3673 (2022). https://doi.org/10.14778/3554821.3554871

[14] Maciej Buze, Thoms E. Woolley, L. Angela Mihai: A Stochastic Framework for Atomistic Fracture. SIAM J. Appl. Math. 82(2): 526-548 (2022). https://doi.org/10.1137/21M1416436

[15] Shan Ma, Matthew J. Woolley, Ian R. Petersen: Synthesis of Linear Quantum Systems to Generate a Steady Thermal State. IEEE Trans. Autom. Control. 67(4): 2131-2137 (2022). https://doi.org/10.1109/TAC.2021.3079291

[16] Jean-Michel Fahmi, Craig A. Woolsey: Port-Hamiltonian Flight Control of a Fixed-Wing Aircraft. IEEE Trans. Control. Syst. Technol. 30(1): 408-415 (2022). https://doi.org/10.1109/TCST.2021.3059928

[17] Marcin Waniek, Tomasz P. Michalak, Michael J. Wooldridge, Talal Rahwan: How Members of Covert Networks Conceal the Identities of Their Leaders. ACM Trans. Intell. Syst. Technol. 13(1): 12:1-12:29 (2022). https://doi.org/10.1145/3490462

[18] Thomas A. Woolman, Philip Lee: Effects of Deep Learning Technologies on Employment in the Field of Digital Communication Systems. Int. J. Innov. Digit. Econ. 12(4): 35-42 (2021). https://doi.org/10.4018/IJIDE.2021100103