Agrica
  • Year: 2023
  • Volume: 12
  • Issue: 2

Vis-NIR spectral differentiation of Albizia procera (Roxb.) Benth. and Saccharum spp.

  • Author:
  • Osmany de la Caridad Aday Díaz1,*, Katia Ojito Ramos2, Rubén Orozco-Morales3, Luís Hernández Santana4
  • Total Page Count: 8
  • Published Online: Feb 2, 2024
  • Page Number: 228 to 235

1Instituto de Investigaciones de la Caña de Azúcar Villa Clara (INICA Villa Clara), Autopista Nacional km 246, Ranchuelo Villa Clara, Cuba, CP:531002

2Departamento de Biología, Facultad de Ciencias Agropecuarias, Universidad Central “Marta Abreu” de Las Villas, Cuba

3Centro de Investigaciones de Métodos Computacionales y Numéricos en la Ingeniería (CIMCNI), Universidad Central “Marta Abreu” de Las Villas, Cuba

4Grupo de Automatización, Robótica y Percepción (GARP), Universidad Central “Marta Abreu” de Las Vilas (UCLV), Cuba

*Corresponding Author : osmany.aday@inicvc.azcuba.cu

Online published on 2 February, 2024.

Abstract

Identification and differentiation of invasive plants in agricultural crops can be done by using appropriate image processing methods. It is first necessary to determine the spectral differences and select those that distinguish the species under study. Albizia procera (Roxb.) Benth., is an invasive species in natural landscapes and agricultural ecosystems such as sugar cane cultivation (Saccharum spp.). The objective of this research was to identify spectral reflectance traits of leaves of A. Procera and Saccharum spp. Fifty leaves of each species were collected. Vis-NIR spectral signatures were acquired at laboratory level with a portable spectrophotometer in the wavelength range 399 to 1697 nm. Large variability of the average features was determined in both species, which is mainly manifested in the amplitude of the reflectance and not in the shape of the spectral characteristic. At 730.48 nm (close to the geometric centre of the red edge), 1032.16 nm and 1395.91 nm, different combinations of the three amplitudes were formed for each species. The results allowed the characterisation of spectral variations and the identification of significant wavelengths from which to proceed to the discrimination and classification of the two crop species studied. It was concluded that proper band selection and local calibration using a spectral classification approach can facilitate the mapping of A. procera in a given geographical area.

Keywords

Precision agriculture, Spectral signature, Weed detection