Search

Accurate information about growing crops allows for regulating the internal stocks of agricultural products and drawing strategies for negotiating agricultural commodities on financial markets. Machine learning methods are widely implemented for crop type recognition and classification based on satellite images. However, field classification is complicated by class imbalance and aggregation of pixel-wise into field-wise forecasting. We propose here a Bayesian methodology for the aggregation of classification results. We report the comparison of class balancing techniques. We also report the comparison of classical machine learning methods and the U-Net convolutional neural network for classifying crops using a single satellite image. The best result for single-satellite-image crop classification was achieved with an overall accuracy of 77.4% and a Macro F1-score of 0.66. Bayesian aggregation for field-wise classification improved the result obtained using majority voting aggregation by 1.5%. We demonstrate here that the Bayesian aggregation approach outperforms the majority voting and averaging strategy in overall accuracy for the single-image crop classification task. [ABSTRACT FROM AUTHOR]

Usage of multispectral satellite imaging data opens vast possibilities for monitoring and quantitatively assessing properties or objects of interest on a global scale. Machine learning and computer vision (CV) approaches show themselves as promising tools for automatizing satellite image analysis. However, there are limitations in using CV for satellite data. Mainly, the crucial one is the amount of data available for model training. This paper presents a novel image augmentation approach called MixChannel that helps to address this limitation and improve the accuracy of solving segmentation and classification tasks with multispectral satellite images. The core idea is to utilize the fact that there is usually more than one image for each location in remote sensing tasks, and this extra data can be mixed to achieve the more robust performance of the trained models. The proposed approach substitutes some channels of the original training image with channels from other images of the exact location to mix auxiliary data. This augmentation technique preserves the spatial features of the original image and adds natural color variability with some probability. We also show an efficient algorithm to tune channel substitution probabilities. We report that the MixChannel image augmentation method provides a noticeable increase in performance of all the considered models in the studied forest types classification problem. [ABSTRACT FROM AUTHOR]