R. Rios-Cabrera1, I. Lopez-Juarez1 and Hsieh Sheng-Jen2
1 Centro de Investigación y de Estudios Avanzados del IPN-Saltillo Grupo de Robótica y Manufactura Avanzada reyes.rios@ieee.org
2 Texas A & M University Engineering Technology and Industrial Distribution Dept.
ABSTRACT
An image processing methodology for the extraction of potato properties is explained. The objective is to determine their quality evaluating physical properties and using Artificial Neural Networks (ANN's) to find misshapen potatoes. A comparative analysis for three connectionist models (Backpropagation, Perceptron and FuzzyARTMAP), evaluating speed and stability for classifying extracted properties is presented. The methodology for image processing and pattern feature extraction is presented together with some results. These results showed that FuzzyARTMAP outperformed the other models due to its stability and convergence speed with times as low as 1 ms per pattern which demonstrates its suitability for real-time inspection. Several algorithms to determine potato defects such as greening, scab, cracks are proposed which can be affectively used for grading different quality of potatoes.
]]> Keywords: ANN, ART theory, pattern recognition, Visual inspection.
RESUMEN
Se explica una metodología para extracción de propiedades de papa usando procesamiento de imágenes. El objetivo es determinar su calidad evaluando propiedades físicas y utilizando Redes Neuronales Artificiales (ANN's) se encuentran deformaciones. Se lleva a cabo un análisis comparativo de tres modelos (Backpropagation, Perceptron y Fuzzy ARTMAP), evaluando velocidad y estabilidad para clasificación de propiedades extraídas. La metodología de procesamiento de imágenes y extracción de características se presenta mostrando algunos resultados. Fuzzy ARTMAP superó a los otros modelos debido a su estabilidad y velocidad de convergencia con tiempos tan bajos como 1 ms por patrón, lo cual demuestra lo apropiado del modelo para inspección en tiempo real. Se proponen varios algoritmos para determinar defectos de la papa tales como color verdoso, manchas, grietas y clasificación de forma, que pueden ser empleados de forma efectiva para clasificar diferentes calidades de papas.
DESCARGAR ARTÍCULO EN FORMATO PDF
REFERENCES
[1]. Liyong Zhou, Vikram Chalana, Yongmin Kim. (1998). "PC-based machine vision system for real-time computer-aided potato inspection," Int. Journal of Imaging Systems and Technology, v 9, n 6, pp. 423-433. [ Links ]
]]>[2]. Brosnan, T. and Sun, D-W. (2002) "Inspection and grading of agricultural and food products by computer vision systems - A review," Computers and Electronics in Agriculture, v 36, n 2-3, 2002, pp. 193-213. [ Links ]
[3]. Brosnan, T. and Sun, D-W. (2004). Improving quality inspection of food products by computer vision - A review, Journal of Food Engineering, v 61, n 1, pp. 3-16. [ Links ]
[4]. J.C. Noordam, G.W. Otten, A.J.M. Timmermans, B.H. van Zwol, (2000)"High speed potato grading and quality inspection based on a colour vision system", Proc. SPIE Vol. 3966, p. 206-217, Machine Vision Applications in Industrial Inspection VIII, Kenneth W. Tobin; Ed. [ Links ]
[5]. Tao, Y., Heinemann, P. H., Varghese, Z., Morrow, C. T., and Sommer, H. J., (1995).. Machine vision for colour inspection of potatoes and apples, Transactions of the ASAE, 38(5), pp. 1555-1561. [ Links ]
[6]. Klein, Lawrence A. (2004), Sensor and Data Fusion, A tool for Information Assessment and Decision Making, SPIE Press, USA. [ Links ]
]]>[7]. Carpenter, G.A. Carpenter, S. Grossberg, N. Markuzon, J.H. Reynolds and D.B. Rosen, (1992), Fuzzy ARTMAP: A neural network architecture for incremental learning of analog multidimensional maps, IEEE Transactions on Neural Networks 3 (5), pp. 698–713. [ Links ]
[8]. Cybenko, G., (1989) Approximation by superposition of a Sigmoidal Function, Mathematics of Control, Signals, and Systems, 2, 303-314. [ Links ]
[9]. Basler Vision Technologies, (2005), Basler A600f user's Manual, http://www.basler-vc.com [ Links ] ]]>