Object Recognition Through Artificial Intelligence Techniques
This paper describes a methodology for object recognition categorized as polyhedron and non-polyhedron. This recognition is achieved through digital image processing combined with artificial intelligence algorithms, such as Hopfield networks. The procedure consists of processing images in search of...
| Main Authors: | , , |
|---|---|
| Format: | Online |
| Language: | eng spa |
| Published: |
Universidad Pedagógica y Tecnológica de Colombia
2020
|
| Subjects: | |
| Online Access: | https://revistas.uptc.edu.co/index.php/ingenieria/article/view/10734 |
| _version_ | 1801706087076855808 |
|---|---|
| author | Ramírez-Arias, José Luis Rubiano-Fonseca, Astrid Jiménez-Moreno, Robinson |
| author_facet | Ramírez-Arias, José Luis Rubiano-Fonseca, Astrid Jiménez-Moreno, Robinson |
| author_sort | Ramírez-Arias, José Luis |
| collection | OJS |
| description | This paper describes a methodology for object recognition categorized as polyhedron and non-polyhedron. This recognition is achieved through digital image processing combined with artificial intelligence algorithms, such as Hopfield networks. The procedure consists of processing images in search of patterns to train the system. The process is carried out through three stages: i) Segmentation, ii) Smart recognition, and iii) Feature extraction; as a result, images of objects are obtained and trained in the designed neuronal network. Finally, Hopfield's network is used to establish the object type as soon as it receives one. The proposed methodology was evaluated in a real environment with a considerable number of detected images; the noisy images recognition uncertainty was 2.6%, an acceptable result considering variable light, shape and color. The results obtained from this experiment show a high recognition level, which represents 97.4%. Out of this procedure, we can assume that it is possible to train new patterns, and it is expected that the model will be able to recognize them. Potentially, the proposed methodology could be used in a vast range of applications, such as object identification in industrial environments, grasping objects using manipulators or robotic arms, tools for blind patients, among other applications. |
| format | Online |
| id | oai:oai.revistas.uptc.edu.co:article-10734 |
| institution | Revista Facultad de Ingeniería |
| language | eng spa |
| publishDate | 2020 |
| publisher | Universidad Pedagógica y Tecnológica de Colombia |
| record_format | ojs |
| spelling | oai:oai.revistas.uptc.edu.co:article-107342021-07-13T02:24:08Z Object Recognition Through Artificial Intelligence Techniques Reconocimiento de objetos a través de técnicas de inteligencia artificial Ramírez-Arias, José Luis Rubiano-Fonseca, Astrid Jiménez-Moreno, Robinson Objects recognition as of 2D images morphologic operations neuronal networks Hopfield network Reconocimiento de imágenes en 2D operaciones morfológicas redes neuronales red de Hopfield This paper describes a methodology for object recognition categorized as polyhedron and non-polyhedron. This recognition is achieved through digital image processing combined with artificial intelligence algorithms, such as Hopfield networks. The procedure consists of processing images in search of patterns to train the system. The process is carried out through three stages: i) Segmentation, ii) Smart recognition, and iii) Feature extraction; as a result, images of objects are obtained and trained in the designed neuronal network. Finally, Hopfield's network is used to establish the object type as soon as it receives one. The proposed methodology was evaluated in a real environment with a considerable number of detected images; the noisy images recognition uncertainty was 2.6%, an acceptable result considering variable light, shape and color. The results obtained from this experiment show a high recognition level, which represents 97.4%. Out of this procedure, we can assume that it is possible to train new patterns, and it is expected that the model will be able to recognize them. Potentially, the proposed methodology could be used in a vast range of applications, such as object identification in industrial environments, grasping objects using manipulators or robotic arms, tools for blind patients, among other applications. En el presente artículo se describe una metodología para la identificación de objetos clasificados en poliedros y no poliedros, este reconocimiento se logra mediante el procesamiento digital de imágenes combinado con el uso de algoritmos de inteligencia artificial, como las redes neuronales de Hopfield. El procedimiento consiste en procesar las imágenes con el fin de obtener los patrones a entrenar, dicho proceso fue desarrollado en tres etapas: i) Segmentación, ii) Reconocimiento inteligente, y iii) Extracción de características; a partir de los resultados obtenidos, en este caso imágenes de los objetos, estos elementos se entrenan en la red neuronal diseñada; finalmente, se usa la red neuronal de Hopfied, la cual, al recibir un nuevo elemento o imagen de un objeto, determinará el tipo de objeto. La metodología propuesta fue evaluada en un ambiente real con un amplio número de imágenes detectadas, la incertidumbre al reconocer imágenes ruidosas, representa el 2.6% de la muestra, ofreciendo una respuesta aceptable frente a condiciones de luz, forma y color variables. Los resultados obtenidos a partir del experimento evidencian un grado alto de reconocimiento, lo cual representa el 97.4 por ciento. A partir de este procedimiento es posible entrenar nuevos patrones con novedosas formas, y se espera que este modelo de reconocimiento sea capaz de reconocer patrones completamente nuevos. La metodología propuesta potencialmente puede tener diferentes aplicaciones, tales como la identificación de objetos en procesos industriales; funciones de agarre de objetos mediante el uso de manipuladores o brazos robóticos; así como en el área de la rehabilitación para ayudar a personas con limitaciones visuales, entre otras. Universidad Pedagógica y Tecnológica de Colombia 2020-04-10 info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion application/pdf application/pdf application/xml https://revistas.uptc.edu.co/index.php/ingenieria/article/view/10734 10.19053/01211129.v29.n54.2020.10734 Revista Facultad de Ingeniería; Vol. 29 No. 54 (2020): Continuos Publication; e10734 Revista Facultad de Ingeniería; Vol. 29 Núm. 54 (2020): Publicación Continua; e10734 2357-5328 0121-1129 eng spa https://revistas.uptc.edu.co/index.php/ingenieria/article/view/10734/9178 https://revistas.uptc.edu.co/index.php/ingenieria/article/view/10734/9177 https://revistas.uptc.edu.co/index.php/ingenieria/article/view/10734/9601 N.A. N.A. Copyright (c) 2020 José Luis Ramírez-Arias, Ph. D., Astrid Rubiano-Fonseca, Ph. D., Robinson Jiménez-Moreno, Ph. D. |
| spellingShingle | Objects recognition as of 2D images morphologic operations neuronal networks Hopfield network Reconocimiento de imágenes en 2D operaciones morfológicas redes neuronales red de Hopfield Ramírez-Arias, José Luis Rubiano-Fonseca, Astrid Jiménez-Moreno, Robinson Object Recognition Through Artificial Intelligence Techniques |
| title | Object Recognition Through Artificial Intelligence Techniques |
| title_alt | Reconocimiento de objetos a través de técnicas de inteligencia artificial |
| title_full | Object Recognition Through Artificial Intelligence Techniques |
| title_fullStr | Object Recognition Through Artificial Intelligence Techniques |
| title_full_unstemmed | Object Recognition Through Artificial Intelligence Techniques |
| title_short | Object Recognition Through Artificial Intelligence Techniques |
| title_sort | object recognition through artificial intelligence techniques |
| topic | Objects recognition as of 2D images morphologic operations neuronal networks Hopfield network Reconocimiento de imágenes en 2D operaciones morfológicas redes neuronales red de Hopfield |
| topic_facet | Objects recognition as of 2D images morphologic operations neuronal networks Hopfield network Reconocimiento de imágenes en 2D operaciones morfológicas redes neuronales red de Hopfield |
| url | https://revistas.uptc.edu.co/index.php/ingenieria/article/view/10734 |
| work_keys_str_mv | AT ramirezariasjoseluis objectrecognitionthroughartificialintelligencetechniques AT rubianofonsecaastrid objectrecognitionthroughartificialintelligencetechniques AT jimenezmorenorobinson objectrecognitionthroughartificialintelligencetechniques AT ramirezariasjoseluis reconocimientodeobjetosatravesdetecnicasdeinteligenciaartificial AT rubianofonsecaastrid reconocimientodeobjetosatravesdetecnicasdeinteligenciaartificial AT jimenezmorenorobinson reconocimientodeobjetosatravesdetecnicasdeinteligenciaartificial |
