ResearchBib Share Your Research, Maximize Your Social Impacts
Sign for Notice Everyday Sign up >> Login

Peaks and Valleys Model for Risks Mitigation in Financial System: A Method Based in Multilevel Thresholding with OBIA for Change Detections in Agricultural Areas, using Remote Sensing

Journal: International Journal of Advanced Engineering Research and Science (Vol.8, No. 10)

Publication Date:

Authors : ;

Page : 060-079

Keywords : Change Detections; Descriptors; LimiariZC; Mitigation; Multilevel Thresholding; Nanosatellites; OBIA; Peaks and Valleys; Python; Remote Sensing.;

Source : Downloadexternal Find it from : Google Scholarexternal


The Geotechnologies has contributed to continuous and agile monitoring, allowing strategic decision, such as monitoring by Remote Sensing in the use and land cover with the agribusiness, which is one of the main sectors of the world economy. Agriculture is one of the responsible for the positive balance of trade in several countries and which many have government policies that subsidize agricultural credits to encourage the sector. Thus, it's necessary to mitigate risks in the financeable agricultural areas, with quick and transparent inspection. In this scenario, a tool in Python language was develop containing a method, called of Peaks and Valleys (PV) Model, for remote monitoring of agricultural production with multitemporal changes detections. The study was in an area in Brazil, using 9 images from the Nanosatellite Planet, from 2017 to 2019. The method has a decision tree that was able to detect changes in the patterns of agricultural areas, issuing assertive signals in cases of deviation in behavior in the remote monitoring of cultivation, from initial cycle, full and final of maturation agricultural, with messages of warning of vegetation growth or alert of loss of vegetation. In model a multilevel thresholding is performed and descriptors extracted (Entropy, Homogeneity, Correlation and Euclidean Distance). The results indicated that when using multilevel thresholding aggregating contextual information with Object-Oriented refinement by Scale descriptor and application of Low Pass Filter by Mean Convolution, there is significant improvement in results. Was possible to assess the quality of the method and its feasibility for remote monitoring in agricultural production, where the model can be used as a significant indicator of oscillation and multitemporal trends in the use and land cover. Thus, the PV Model can facilitate inspection by of the countries with subsidies for agriculture, whether by inspectors from the Government or by Financial Institutions, in addition to reducing costs in the operational process concentrating face-to-face visits only for areas of large hectares.

Last modified: 2021-10-30 14:59:20