Efecto de los filtros morfológicos en los procesos de detección de objetos en movimiento

Sergio Alexander Castro-Casadiego; Karla Yohana Sánchez-Mojica; Karla Cecilia Puerto-López; Carlos Vicente Niño-Rondón; Byron Medina-Delgado; Dinael Guevara-Ibarra

doi:10.61799/2216-0388.676

PDF (Spanish)

Effect of morphological filters on the moving object detection process

Sergio Alexander Castro-Casadiego

Universidad Francisco de Paula Santander

Karla Yohana Sánchez-Mojica

Fundación de Estudios Superiores Comfanorte- Fesc

Karla Cecilia Puerto-López

Universidad Francisco de Paula Santander

Carlos Vicente Niño-Rondón

Universidad Francisco de Paula Santander

Byron Medina-Delgado

Universidad Francisco de Paula Santander

Dinael Guevara-Ibarra

Universidad Francisco de Paula Santander

DOI: https://doi.org/10.61799/2216-0388.676

Published

2021-01-01

Issue

Vol. 11 No. 21 (2021)

Section

Artículo Originales

License

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

How to Cite

[1]

Castro-Casadiego, S.A. et al. 2021. Effect of morphological filters on the moving object detection process. Mundo FESC Journal. 11, 21 (Jan. 2021), 87–95. DOI:https://doi.org/10.61799/2216-0388.676.

Download Citation

Dimensions Metrics

https://plu.mx/plum/a/?doi=10.61799/2216-0388.676

Keywords:
Abstract

back ground subtraction, morphological filters, objectdetection, image processing

In the processes of background subtraction applied to the detection of moving objects, one of the most relevant stages is the filtering by morphology, where the image is simplified and most of the shape characteristics of the objects are preserved. Therefore, a comparison is made between the operations of dilation, erosion, opening, closing and gradient in video images with static background, where people circulate in uncontrolled environments, to determine their behavior in the detection and counting of people. Image processing is performed in Python language and the specialized computer vision package OpenCV is used. In addition, a graphical user interface was developed using Tkinter to enter the values of the size and shape of the structural element for processing. When applying the morphological filtering by dilatation, a success in the detections of 82.28 %, with erosion the accuracy was 81.86 %, while, with the opening, closing and gradient operations the accuracy was 83.69 %, 93.07 % and 87.69 % respectively.

Effect of morphological filters on the moving object detection process

Published

Issue

Section

License

How to Cite

Dimensions Metrics

Keywords:

Abstract

Most read articles by the same author(s)

Language