Saeid Ahmadi; Hadiseh Hasani
Abstract
Nowaday, there are wide applications for satellite images in agriculture monitoring and management. According to high spatial, spectral and temporal resolution of Sentinel-2 images, we used them for precise agriculture in Qorveh country. Proposed methd consist of five step: firstly, multi-temporal images ...
Read More
Nowaday, there are wide applications for satellite images in agriculture monitoring and management. According to high spatial, spectral and temporal resolution of Sentinel-2 images, we used them for precise agriculture in Qorveh country. Proposed methd consist of five step: firstly, multi-temporal images are collected based on agriculture calender of crops. Then feature space is generated based on spectral reflectance and vegetation indices which consists of 70 features. According to high dimensionality of feature space, principle component analysis is applied to reduce its dimension. Four power classifiers include support vector machine, k-nearest neighbour, multi-layer perceptron and random forests classify the reduced spectral feature space. On the other hand, spatial information are extracted from multi-temporal multispectral images. For this pupose, strandard deviation (STD) maps are generated for red, NIR and SWIR bands of each epoch. Then, by averaging the STD maps, final STD map is obtained. Edge detection is performed on STD map and it improves by removing small lines, smoothing, thining, etc. Finally, crop mapping is done by fusion of four classification maps and agriculture farm boundaries. The obtained results show that classification accuracy of k-nearest neighbour, support vector machine, multi-layer perceptron and random forest classifiers are 77.78%, 79,16%, 76.41% and 76.89%, respectively. The overall accuracy of the proposed method improve up to 94.72% which proves high potential of the proposed method.
Zeinab Ghodsi; Mir Masoud Kheirkhah Zarkesh; Bagher Ghermezcheshmeh
Abstract
Land-cover/land-use maps are necessary for monitoring land changes and proper planning for managers in agriculture, natural resources and environment fields each year. The method of field data collection using GPS and land survey is time-consuming and costly. Therefore satellite images which have entire ...
Read More
Land-cover/land-use maps are necessary for monitoring land changes and proper planning for managers in agriculture, natural resources and environment fields each year. The method of field data collection using GPS and land survey is time-consuming and costly. Therefore satellite images which have entire coverage and repetition of collection, low cost and real-time data, are usually used so that land-cover/land-use maps are produced. Accurate mapping using technique suitable for today is a key factor. Although in the past, conventional classification methods have been applied to images such as Landsat, using new satellite images and modern classifiers specially machine learning has been growing recently and their effectiveness in preparing land-cover/land-use maps has been very successful. Another advantage of satellite images is repetitious collection and according to that, vegetation changes through time can be used to differentiate land cover types. The Sentinel-2 satellite with the superiority of a pixel rating of 10 meters is one of the appropriate tools to discriminate land cover types. In the current study, Support Vector Machine and Random Forest classifiers on multi-temporal Sentinel-2 images were used to differentiate land use and crop types of Sanjabi plain in Ravansar and their accuracies were compared. To do so, after sampling, Principal Component Analysis was performed for four dates in crops’ growing season and PC1,2,3 bands of the images were combined. The two techniques were implemented on the layerstacks of PC1,2,3 bands of the images and the training samples. Results of accuracy assessments showed that Support Vector Machine, with overall accuracy of 91.36% and Kappa coefficient of 0.8927, produces a more precise land use and crop map rather than Random Forest method.
