Geology, geography and global energy

Scientific and Technical Journal

Application of remote sensing data to evaluate and classify crop conditions

2012. №4, pp. 175-180

Pogorelov Anatoliy V. - D.Sc. in Geography, Professor, Kuban State University, 149 Stavropolskaya st., Krasnodar, Russian Federation, 350075, pogorelov@nm.ru

Kuznetsov Konstantin V. - Post-graduate student, Kuban State University, 149 Stavropolskaya st., Krasnodar, Russian Federation, 350075, kvkuznetsov88@gmail.com

The article covers the application of remote sensing data to evaluate and classify crop conditions. The data in the proposed method could, for example, be used to efficiently assess the condition of winter wheat fields in Krasnodar, as well as the area defined by its broader regional boundaries. An agricultural area, Krasnodar is currently among the leading producers of much agricultural produce in the Russian Federation. It already uses real-time monitoring of non-residential land and crops as a tool to facilitate further agricultural development. However, modern methods of gauging distances, such as remote sensing, need to be used to trace the region’s crops, due to its large annual acreage edge (more than 3.6 million hectares). In crop monitoring, one of the priority tasks is edge detection of ’hot spots’ and definition of their coverage areas. By hot spots, the document is referring to satellite imagery of local areas in a field of crops having significant heterogeneity. The data from the imagery needs to be applied in a timely fashion at an early stage of the culture’s maturation to compensate for local challenges such as winter reseeding. In the critique’s view, the data accumulated on the distribution of these sites could also be used to establish the causes of potential problems (natural or man-made). The study has suggested using a series of techniques aimed at estimating the local spatial uniformity among crops to isolate areas with hot spots. One of the local methods for detecting defective crop areas is called ’fuzzy classification;’ It involves, inter alia, carrying out learning-oriented classes with potentially uniform elements that are significantly different from other classes of elements. In the course of its discussion, the research paper also touches on some theoretical aspects of the proposed classification methodology.

Key words: fuzzy classification,agriculture,satellite imagery,vegetation index,GIS (geographic information system) Modeling

Read