Search

Road segmentation from aerial images is a challenging yet crucial task, underpinning significant applications in urban planning, navigation, and transportation systems. In this study, we employ a conditional Generative Adversarial Network (GAN) architecture that synergistically integrates the strengths of Attention U-Net and PatchGAN to address this task. The Attention U-Net, serving as the generator, is trained on the publicly available Massachusetts Roads dataset, with an emphasis on critical regions while concurrently disregarding the irrelevant ones, thereby enhancing the accuracy of road segmentation. Simultaneously, the PatchGAN discriminator ensures the generation of sharp, high-quality segmentations. Through this cooperative approach, we have achieved an overall accuracy of 98.2%, a recall of 82.30%, a precision of 78.66%, an Intersection over Union (IoU) of 67.19%, and an F1 score of 80.44% on our dataset. While these results are promising, they also highlight areas for improvement, particularly in reducing false positives and enhancing the identification of all road pixels, underscoring the potential for future refinements in this research domain.

Land cover classification is a vital application area in the satellite image processing domain. Texture is a useful feature in land cover classification. In this paper, we propose a distributed texture-based land cover classification algorithm using Hidden Markov Model (HMM). Here, HMM is used for texture-based classification of remotely sensed images. Furthermore, to enhance the performance, data-intensive remotely sensed image is segmented and distributed into parallel sessions. Experiments were conducted on IRS P6 LISS-IV data, and the results were evaluated based on the confusion matrix, classification accuracy, and Kappa statistics. These results indicate that the proposed algorithm achieves a classification accuracy of 88.75%.

Texture is a significant spatial property that precisely captures patterns in a satellite image. Texture-based classification plays a vital role in land use-land cover application of remotely sensed images. In this paper, texture features were extracted using Multivariate Local Binary Pattern (MLBP), Multivariate Local Texture Pattern (MLTP), Multivariate Advanced Local Binary Pattern (MALBP), wavelet and Gabor wavelet. Texture-based classification was performed on IRS-P6, LISS-IV data, and the results were evaluated based on error matrix, classification accuracy and kappa statistics. From the experiments, it was found that MLTP outperformed other texture models.

Texture features play a vital role in land cover classification of remotely sensed images. Local binary pattern (LBP) is a texture model that has been widely used in many applications. Many variants of LBP have also been proposed. Most of these texture models use only two or three discrete output levels for pattern characterization. In the case of remotely sensed images, texture models should be capable of capturing and discriminating even minute pattern differences. So a multivariate texture model is proposed with four discrete output levels for effective classification of land covers. Remotely sensed images have fuzzy land covers and boundaries. Support vector machine is highly suitable for classification of remotely sensed images due to its inherent fuzziness. It can be used for accurate classification of pixels falling on the fuzzy boundary of separation of classes. In this work, texture features are extracted using the proposed multivariate descriptor, MDLTP/MVAR that uses multivariate discrete local texture pattern (MDLTP) supplemented with multivariate variance (MVAR). The classification accuracy of the classified image obtained is found to be 93.46 %.