Memoria Temporal Jerárquica Bioinspirada para la Clasificación Automatizada de Fibras de Llama en Entornos Adversos y con Datos Limitados.
DOI:
https://doi.org/10.33414/rtyc.56.40-58.2026Palabras clave:
Memoria Temporal Jerárquica (HTM), Fibras de Llama, Representaciones Distribuidas Dispersas (SDR)Resumen
El presente estudio propone una red de Memoria Temporal Jerárquica (HTM - Hierarchical Temporal Memory) para la clasificación automatizada de fibras de llama (Lama glama) basada en patrones de medulación (no medulada, continua, fragmentada o interrumpida). El marco se fundamenta en los principios neurobiológicos de la neocorteza. Aprovecha las Representaciones Distribuidas Dispersas (SDR - Sparse Distributed Representations) y el aprendizaje hebbiano adaptativo para adquirir características morfológicas complejas a partir de conjuntos de datos limitados. La evaluación demostró que la red HTM alcanzó una precisión de clasificación de 0,942 y una puntuación F1 de 0,941, superando el rendimiento de los modelos de Red Neuronal Convolucional (0,903) y Máquina de Vectores de Soporte (0,877). El modelo propuesto también exhibió capacidad de generalización y robustez ante perturbaciones de entrada (manteniendo una puntuación F1 >0,85 con un 20 % de ruido), junto con eficiencia computacional (15 ms por inferencia). Estos resultados validan la eficacia de la arquitectura HTM para mitigar las limitaciones críticas de escasez de datos y escalabilidad limitada. En consecuencia, este trabajo subraya el potencial del modelo para su implementación en aplicaciones de campo con condiciones de adquisición de datos subóptimas, lo que representa un avance significativo en la inteligencia artificial eficiente en el uso de recursos para el análisis textil especializado.
Descargas
Citas
Ahmad, S., & Hawkins, J. (2015). Properties of sparse distributed representations and their application to hierarchical temporal memory. https://arxiv.org/abs/1503.07469
Amorena, J. I., Álvarez, D. M. E. & Fernández-Ahumada, E. (2021). Development of Calibration Models to Predict Mean Fibre Diameter in Llama (Lama glama) Fleeces with Near Infrared Spectroscopy. Animals 11(7), 1998. https://doi.org/10.3390/ani11071998
Azémard, C., Dufour, E., Zazzo, A., Wheeler, J. C., Goepfert, N., Marie, A., & Zirah, S. (2021). Untangling the fibre ball: Proteomic characterization of South American camelid hair fibres by untargeted multivariate analysis and molecular networking. Journal of proteomics, 231, 104040. https://doi.org/10.1016/j.jprot.2020.104040
Barra Novoa, R. (2023). Analysis of Competitiveness, Market and Consumer Trends for the Camelid Livestock Value Chain. Market and Consumer Trends for the Camelid Livestock Value Chain (July 31, 2023). https://hal.science/hal-04353403v1
Barra Novoa, R. (2024). Analysis of Market Trends and Competitiveness of the Camelid Livestock Value Chain in the North of Chile. https://doi.org/10.20944/preprints202410.0196.v1
Bourne, R. (2010). Contrast adjustment. In: Fundamentals of digital imaging in medicine. 109-135. Springer, London. https://doi.org/10.1007/978-1-84882-087-6_6
Condor, J. R., Paucar, Y., & Paucar, R. (2022). Caracterización morfológica, parámetros productivos y características textiles en llamas (Lama glama) K’ara en Huancavelica. Revista de investigación Agropecuaria Science and Biotechnology, 2(3), 21-31. https://doi.org/10.25127/riagrop.20223.846
Delbón, N., Machado, A. S., Cabrera, V. A., Wiemer, A. P., Matesevach Becerra, A. M., Chiarini, F. E., & Saur Palmieri, V. (2024). Técnicas de histología vegetal: Manual teórico-práctico. ISBN: 978-631-00-3158-3. https://ri.conicet.gov.ar/handle/11336/253125
Cui, Y., Ahmad, S., & Hawkins, J. (2017). The HTM spatial pooler—A neocortical algorithm for online sparse distributed coding. Frontiers in computational neuroscience, 11, 111 https://doi.org/10.3389/fncom.2017.00111
Fallas-Moya, F., & Torres-Rojas, F. (2018). Object recognition using hierarchical temporal memory. In Intelligent Computing Systems: Second International Symposium, ISICS 2018, Merida, Mexico, March 21-23, 2018, Proceedings 2 (pp. 1-14). Springer International Publishing. https://doi.org/10.1007/978-3-319-76261-6_1
Frank, E. (2017). Comercialización de Fibras de Camélidos Sudamericanos. Serie Documentos Internos SUPPRAD 5, Red SUPPRAD 2016. https://hdl.handle.net/20.500.12032/44260
Frank, E., Prieto, A., Castillo, M. F., Segueti Frondizi, D. G., Mamani-Cato, R. H. & Hick, M. V. H. (2021). The Prickle Effect Comes From Fabrics Made of South American Camelid (Alpaca and Lama) Fibers. Mechanical and/or Genetic Solutions. A Review. European Journal of Applied Sciences 9(3). https://doi.org/10.14738/aivp.93.10130
Hao, X., Liu, L., Yang, R., Yin, L., Zhang, L., & Li, X. (2023). A review of data augmentation methods of remote sensing image target recognition. Remote Sensing, 15(3), 827. https://doi.org/10.3390/rs15030827
Hawkins, J., & Ahmad, S. (2016). Why neurons have thousands of synapses, a theory of sequence memory in neocortex. Frontiers in neural circuits, 10, 23. https://doi.org/10.3389/fncir.2016.00023
Hawkins, J., & Blakeslee, S. (2004). On intelligence: how a new understanding of the brain will lead to the creation of truly intelligent machines. Macmillan. ISBN: 9780805074567 / 0805074562
Hawkins, J., George, D. (2006) Hierarchical temporal memory, concepts, theory, and terminology. Numenta, Tech. Rep.
Hawkins, J., George, D. & Niemasik, J. (2009). Sequence memory for prediction, inference and behaviour. Philosophical Transactions of the Royal Society B: Biological Sciences 364(1521). 1203–1209. https://doi.org/10.1098/rstb.2008.0322
Ibrayev, T., Myrzakhan, U., Krestinskaya, O., Irmanova, A., & James, A. P. (2018). On-chip face recognition system design with memristive hierarchical temporal memory. Journal of Intelligent & Fuzzy Systems: Applications in Engineering and Technology, 34(3), 1393-1402. https://doi.org/10.3233/JIFS-169434
Kishore, A., Pal, B., & Sarkar, P. (2024). Camelids for sustainability: a socio-economic perspective. Asian J. Environ. Ecol, 23(1), 53-72. https://doi.org/10.9734/AJEE/2024/v23i1521
Khobragade, V., Nirmal, J., & Chedda, S. (2022). Revaluating Pretraining in Small Size Training Sample Regime. International Journal of Electrical and Electronics Research, 10(3), 694–704. https://doi.org/10.37391/ijeer.100346
Lamas, H. (2007). Desarrollo del encadenamiento productivo de la llama en la provincia de Jujuy, República Argentina. Primer Borrador de Avance, proyecto “Desarrollo del encadenamiento productivo de la llama en la provincia de Jujuy, República Argentina”.
Lu, K., Luo, J., Zhong, Y., & Chai, X. (2019). Identification of wool and cashmere SEM images based on SURF features. Journal of engineered fibers and fabrics, 14, https://doi.org/10.1177/1558925019866121
Luo, J., Lu, K., Chen, Y., & Zhang, B. (2021). Automatic identification of cashmere and wool fibers based on microscopic visual features and residual network model. Micron, 143, 103023. https://doi.org/10.1016/j.micron.2021.103023
Malik, N., & Bzdok, D. (2022). From YouTube to the brain: Transfer learning can improve brain-imaging predictions with deep learning. Neural Networks, 153, 325–338. https://doi.org/10.1016/j.neunet.2022.06.014
Mamani-Cato, R. H., Frank, E. N., Prieto, A., Castillo, M. F., Condori-Rojas, N., & Hick, M. V. H. (2022). Effect of fibre diameter, prickle factor and coarse fibre bias on yarn surface hairiness in South American camelids (SAC) fibre. Fibers, 10(2), 18. https://doi.org/10.3390/fib10020018
McGregor, B. A. (2018). Physical, Chemical, and Tensile Properties of Cashmere, Mohair, Alpaca, and Other Rare Animal Fibers. In: Handbook of Properties of Textile and Technical Fibres, 2nd ed.; Bunsell, A.R., Ed.; Woodhead Publishing: Sawston, UK. pp. 105–136. ISBN 978-0-08-101272-7. https://doi.org/10.1016/B978-0-08-101272-7.00004-3
Miconi, T. (2021). Hebbian learning with gradients: Hebbian convolutional neural networks with modern deep learning frameworks. arXiv preprint arXiv:2107.01729. https://doi.org/10.48550/arXiv.2107.01729
Mnatzaganian, J., Fokoué, E., & Kudithipudi, D. (2016). A Mathematical Formalization of Hierarchical Temporal Memory’s Spatial Pooler. arXiv. https://doi.org/10.48550/arxiv.1601.06116
Mueller, J. P., Rigalt, F., Lamas, H., Sacchero, D. M., Cancino, A. K. & Wurzinger, M. (2015). Fibre quality of South American camelids in Argentina: a review. Animal Genetic Resources. 2015, 56, 97–109. https://doi.org/10.1017/S2078633614000496
Musa, P., Al Rafi, F., & Lamsani, M. (2018). A Review: Contrast-Limited Adaptive Histogram Equalization (CLAHE) methods to help the application of face recognition. In: 2018 third international conference on informatics and computing (ICIC) (pp. 1-6). IEEE. https://doi.org/10.1109/IAC.2018.8780492
Ofir, N., Galun, M., Alpert, S., Brandt, A., Nadler, B., & Basri, R. (2019). On detection of faint edges in noisy images. IEEE transactions on pattern analysis and machine intelligence, 42(4), 894-908. https://doi.org/10.1109/TPAMI.2019.2892134
Pietroń, M., Wielgosz, M., & Wiatr, K. (2016). Formal analysis of HTM spatial pooler performance under predefined operation conditions. In Rough Sets: International Joint Conference, IJCRS 2016, Santiago de Chile, Chile, October 7–11, 2016, Proceedings (pp. 396-405). Springer International Publishing. https://doi.org/10.1007/978-3-319-47160-0_36
Pinares, R., & Quispe, E. (2024). Características de las fibras meduladas en alpaca huacaya y suri de Perú. Chilean journal of agricultural & animal sciences, 40(2), 332-340. https://doi.org/10.29393/CHJAAS40-28CCPQ20028
Quispe, D., Castillo, P., Yana, W., Vilcanqui, H. & Apaza, E. (2024). Características Textiles de la Fibra de Alpacas Suri de la Feria Ganadera del Sur del Perú. Chilean journal of agricultural & animal sciences, 40(1), 178-189. https://dx.doi.org/10.29393/chjaas40-17ctde50017
Quispe, M., Trigo, J. D., Serrano-Arriezu, L., Huere, J., Quispe, E., & Beruete, M. (2025). Classification of South American Camelid and goat fiber samples based on fourier transform infrared spectroscopy and machine learning. The Journal of The Textile Institute, 116(2), 198-207. https://doi.org/10.1080/00405000.2024.2324209
Rubiagatra, D., Wibawa, A. D., Lejap, M. Y. L., Pratama, B. G., & Oktavian, R. (2023). Gabor filter and Canny edge detection for ear biometrics identification. In: 2023 International Seminar on Intelligent Technology and Its Applications (ISITIA) (pp. 564-569). IEEE. https://doi.org/10.1109/ISITIA59021.2023.10220442
Sarmiento-Ramos, J. L. (2020). Aplicaciones de las redes neuronales y el deep learning a la ingeniería biomédica. Revista UIS Ingenierías, 19(4), 1-18. https://doi.org/10.18273/revuin.v19n4-2020001
Streat, L., Kudithipudi, D., & Gomez, K. (2016). Non-volatile Hierarchical Temporal Memory: Hardware for Spatial Pooling. arXiv. https://doi.org/10.48550/arxiv.1611.02792
Vilá, B., & Arzamendia, Y. (2022). South American Camelids: their values and contributions to people. Sustainability Science, 17(3), 707-724. https://doi.org/10.1007/s11625-020-00874-y
Wang, F., & Jin, X. (2018). The application of mixed-level model in convolutional neural networks for cashmere and wool identification. International Journal of Clothing Science and Technology, 30(5), 710-725. https://doi.org/10.1108/IJCST-11-2017-0171
Xing, W., Deng, N., Xin, B., Liu, Y., Chen, Y., & Zhang, Z. (2019b). Identification of extremely similar animal fibers based on matched filter and HOG-SVM. IEEE Access, 7, 98603-98617. https://doi.org/10.1109/ACCESS.2019.2923225
Xing, W., Liu, Y., Deng, N., Xin, B., Wang, W., & Chen, Y. (2020). Automatic identification of cashmere and wool fibers based on the morphological features analysis. Micron, 128, 102768. https://doi.org/10.1016/j.micron.2019.102768
Xing, W., Liu, Y., Xin, B., Zang, L., & Deng, N. (2022). The application of deep and transfer learning for identifying cashmere and wool fibers. Journal of Natural Fibers, 19(1), 88-104 https://doi.org/10.1080/15440478.2020.1727817
Xing, W., Xin, B., Deng, N., Chen, Y., & Zhang, Z. (2019a). A novel digital analysis method for measuring and identifying of wool and cashmere fibers. Measurement, 132, 11-21. https://doi.org/10.1016/j.measurement.2018.09.032
Yildiz, K. (2020). Identification of wool and mohair fibres with texture feature extraction and deep learning. IET Image Processing, 14(2), 348-353. https://doi.org/10.1049/iet-ipr.2019.0907
Zang, L., Xin, B., & Deng, N. (2022a). Identification of overlapped wool/cashmere fibers based on multi-focus image fusion and convolutional neural network. Journal of Natural Fibers, 19(13), 6715-6726. https://doi.org/10.1080/15440478.2021.1932669
Zang, L., Xin, B., & Deng, N. (2022b). Identification of wool and cashmere fibers based on multiscale geometric analysis. The Journal of The Textile Institute, 113(6), 1001-1008. https://doi.org/10.1080/00405000.2021.1914399
Zhong, Y., Lu, K., Tian, J., & Zhu, H. (2017). Wool/cashmere identification based on projection curves. Textile Research Journal, 87(14), 1730-1741. https://doi.org/10.1177/0040517516658516
Zhu, Y., Zhao, L., Chen, X., Li, Y., & Wang, J. (2023a). Animal fiber recognition based on feature fusion of the maximum inter-class variance. AUTEX Research Journal, 23(4), 560-566. https://doi.org/10.2478/aut-2022-0031
Zhu, Y., Zhao, L., Chen, X., Li, Y., & Wang, J. (2023b). Identification of cashmere and wool based on LBP and GLCM texture feature selection. Journal of Engineered Fibers and Fabrics, 18. https://doi.org/10.1177/15589250221146548
Zhu, Y., Wang, X., Gu, M., Hu, G., & Li, W. (2024). Application of unsupervised feature selection in cashmere and wool fiber recognition. Journal of Natural Fibers, 21(1), Article 2311306. https://doi.org/10.1080/15440478.2024.2311306
Zoccola, M., Bhavsar, P, Anceschi, A. & Patrucco, A. (2023). Analytical Methods for the Identification and Quantitative Determination of Wool and Fine Animal Fibers: a review. Fibers 2023, 11, 67. https://doi.org/10.3390/fib11080067
Zúñiga, E. A., Olarte Daza, C. U., & Quispe Coaquira, J. E. (2022). Calidad textil de la fibra descerdada de llamas (Lama glama) en piso ecológico húmedo. RIIARn: Revista de Investigación e Innovación Agropecuaria y de Recursos Naturales, 9(3), 68-78. https://doi.org/10.53287/hfvl7514as10k
Zyarah, A. M., & Kudithipudi, D. (2023). An Overview of the Hierarchical Temporal Memory Accelerators. In: Artificial Intelligence Applications and Reconfigurable Architectures, A.D. Thakare and S.U. Bhandari (eds.), 63-94. https://doi.org/10.1002/9781119857891.ch4
Descargas
Publicado
Cómo citar
Número
Sección
Licencia
Derechos de autor 2026 Marcelo Arcidiácono, Dolores María Eugenia Álvarez
Esta obra está bajo una licencia internacional Creative Commons Atribución-NoComercial 4.0.
