A novel approach for optimal weight factor of DT-CWT coefficients for land cover classification using MODIS data.

Abstract : Presently, there is a need to explore the possibility to maximize the use of MODIS (Moderate Resolution Imaging Spectroradiometer) data as it has very good spectral (36 bands) and temporal resolution whereas its spatial resolution is moderate i.e. 250m, 500m, and 1km. Because of its moderate spatial resolution, its application for land cover classification is limited. Therefore, in this paper, an attempt has been made to enhance its spatial resolution and utilize the information contained in the different bands together to achieve good land cover classification accuracy, so that, in future, MODIS data can be used more effectively. For resolution enhancement, modified dual tree complex wavelet transform (DT-CWT) has been employed, where DT-CWT has been modified by critically analyzing the effect of weight factor of the DT-CWT coefficients on land cover classification. For this purpose, image statistics parameter like Mean of the image has also been considered. The proposed technique has been applied on the six bands of MODIS data which have spatial resolution of 500m. It is observed that weight factor of the high-frequency sub-bands is quite sensitive for computation of classification accuracy. Index Terms— DT-CWT, Resolution enhancement, wavelets, weights, MODIS 1.INTRODUCTION Satellite images are being used in various applications such as geoscience studies, astronomy and geographical information systems where their resolution plays a critical role but on the other hand, directly obtaining a high resolution data is an another herculean task because of high cost of sensor. Land cover classification from satellite data is a central topic in satellite imaging applications. Therefore, it becomes a necessity to develop and utilize a reliable resolution enhancement technique to obtain accurate information as much as possible as per application from the freely available moderate resolution satellite data. In this regard, many image resolution enhancement techniques have been developed which are interpolations (nearest neighbor, bilinear and bicubic) and wavelets (DWT, SWT, WZP etc.) based. Interpolation techniques [1] have been widely used for resolution enhancement but it results in loss of edges (i.e., high frequency components) of an image. Nowadays, resolution enhancement is being carried out in the wavelet domain. There are many wavelet transforms which have acquired the place. Discrete wavelet transform (DWT) [2] has also been widely used in order to preserve the high-frequency components of the image but its disadvantage is that it ends up with some ringing artifacts into the image since it is not found to be shift-invariant because of decimations and suppression of wavelet coefficients exploited by DWT. It basically suffers from four shortcomings i.e., oscillations, shift variance, aliasing and lack of directionality which can lead to some artifacts in the image and difficulties in signal modeling. Hence, the DWT has somewhat disappointed the researchers for satellite images. Therefore, in order to alleviate all these drawbacks of DWT [2,3], a new kind of wavelet was introduced by Kingsbury which is known as DT-CWT (Dual tree complex wavelet transform) [1,3]. It possesses shift-invariant property and has the capability of improving directional resolution (because of good directional sensitivity) as compared to that of the decimated DWT. That's why, DT-CWT has been employed in this paper for resolution enhancement of moderate resolution satellite images. It is foremost to discover the possibility of maximizing the use of freely available satellite data like MODIS. It consists of several bands in which different information is present, but has certain limitations as well like low spatial resolution i.e. 500m which is a major obstacle in obtaining that information accurately. Many researchers have worked on resolution enhancement techniques for visualization enhancement whereas in this paper, main motive is to enhance the land cover classification accuracy which is not reported much for satellite images like MODIS yet. Variance minimization [4] has also been explored by several researchers for weights optimization but it is somewhat 4528 978-1-5090-3332-4/16/$31.00
Type de document :
Communication dans un congrès
Geoscience and Remote Sensing Symposium (IGARSS), 2016 IEEE International, Jul 2016, Beijing China. Geoscience and Remote Sensing Symposium (IGARSS), 2016 IEEE International, 2016, 〈http://ieeexplore.ieee.org/abstract/document/7730181/〉. 〈10.1109/IGARSS.2016.7730181〉
Liste complète des métadonnées

https://hal.inria.fr/hal-01400795
Contributeur : H. Yahia <>
Soumis le : mardi 22 novembre 2016 - 14:48:02
Dernière modification le : mercredi 3 janvier 2018 - 14:18:08
Document(s) archivé(s) le : mardi 21 mars 2017 - 07:50:55

Fichier

garg-et-al-igarss-2016.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Collections

Citation

Akanksha Garg, Sashi Vardhan Naidu, Shruti Gupta, Dharmendra Singh, Nicolas Brodu, et al.. A novel approach for optimal weight factor of DT-CWT coefficients for land cover classification using MODIS data.. Geoscience and Remote Sensing Symposium (IGARSS), 2016 IEEE International, Jul 2016, Beijing China. Geoscience and Remote Sensing Symposium (IGARSS), 2016 IEEE International, 2016, 〈http://ieeexplore.ieee.org/abstract/document/7730181/〉. 〈10.1109/IGARSS.2016.7730181〉. 〈hal-01400795〉

Partager

Métriques

Consultations de la notice

216

Téléchargements de fichiers

47