Registration techniques for multiespectral images collected by a mobile terrestrial system.

• Grant number: 23/15204-2
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: November 01, 2023
• End date: October 31, 2024
• Field of knowledge: Physical Sciences and Mathematics - Geosciences - Geodesy
• Principal Investigator: Antonio Maria Garcia Tommaselli
• Grantee: Beatriz Coêlho Silva
• Host Institution: Faculdade de Ciências e Tecnologia (FCT). Universidade Estadual Paulista (UNESP). Campus de Presidente Prudente. Presidente Prudente , SP, Brazil
• Associated research grant: 21/06029-7 - High resolution remote sensing for digital agriculture, AP.TEM

Abstract

The monitoring of agricultural crops can be carried out with the support of mobile mapping systems, which combine geodetic positioning technologies with mobile sensors, allowing the acquisition of data dynamically, at a low cost. The use of a low-cost multispectral mobile sensor, such as the cameras of the Israeli company Agrowing, is an alternative for the study of cultivation areas by proximal remote sensing. In this work, an Agrowing Sony Alpha 7R2 Sextuple Multispectral camera will be used, which is a model that combines a commercial digital camera with a special lens with six lenses and filters, which allows the acquisition of the image in 14 spectral bands, between 400 nm and almost 900 nm, which can be recorded and extracted using the company's own software. The algorithms implemented for the recording use translations only, which ensure a correct co-registration between the bands for the air case. However, in the terrestrial case, the object and the sensor are then very close together and the separation between the lenses causes a parallax effect, depending on the depth of the object. To solve this problem, logging should be done depth-dependent or pixel-by-pixel. Therefore, it is necessary to study other techniques. In this project it is intended to study ways of recording the spectral bands for the terrestrial case with images collected at close range. Initially, it is necessary to characterize the parallaxes between the quadrants of the image, caused by the small difference in position between the perspective centers of the six heads, causing six different points of view of each scene.