Object-Based Crop Mapping Using Multi-Temporal Landsat 8 Imagery

Abstract

This study aimed to map crop pattern using object-based classification technique within a Mediterranean agricultural land in Turkey. Many Mediterranean land covers show similar spectral characteristics that make it difficult to identify in feature space by simple per-pixel classifiers. Therefore, an object-based classification is a potential solution for the classification of land cover in such environments., Appropriate segmentation parameters play a vital role for accurate mapping in object-based image classification. The optimum segmentation parameters were determined by testing on complex agricultural parcels. Each agricultural parcel was assigned a crop type using object-based classification techniques. Two different periods, classified as winter and summer, were classified with high accuracy in the study area. For the crop mapping, March and April are the best imaging times for winter, and between May and August are the best imaging times for summer crop season. Overall accuracy of the classification results, were derived through kappa statistics of 0.85 for winter and summer crop mapping.

Authors and Affiliations

Ahmet Cilek, Suha Berberoglu

Keywords

Related Articles

Review Study of soil behavior mix with waste Plastic

Infrastructure is a major sector that propels overall development of Indian economy. The foundation is very important for any structure and it has to be strong enough to support the entire structure. For foundation to be...

An Area Efficient and High Speed Reversible Multiplier Using NS Gate

In digital computer system a major problem has been found that the Power dissipation which leads to bring some research on the methods to decrease this Area efficient, high speed. This is the main cause to give birth to...

Impact of Initialization shapes on the Performance ofWaveform Optimization

This paper aims to examine the effect of initialization for the shape of waveform that can be used in multicarrier systems such asOrthogonal Frequency-Division Multiplexing (OFDM) and Filter Bank Multi-Carrier (FBMC)for...

A Study of Social Media Data and Data Mining Techniques

Artificial Neural Networks (ANNs) has highly interconnected elements (neurons) which unanimously work to solve the specific problems. Recently ANNs are involved in the areas like image and speech recognition, character &...

Effects of Hemp Flour, Seeds And Oil Additions on Bread Quality

The minerals, dietary fibers and essential fatty acids of several hemp products (flour, seeds and oil) represent valuable compounds sources that could transform the bread in a functional food. The aim of this work was ev...

Download PDF file
  • EP ID EP393970
  • DOI 10.9790/9622-0804023437.
  • Views 106
  • Downloads 0

How To Cite

Ahmet Cilek, Suha Berberoglu (2018). Object-Based Crop Mapping Using Multi-Temporal Landsat 8 Imagery. International Journal of engineering Research and Applications, 8(4), 34-37. https://europub.co.uk/articles/-A-393970