Our society is getting more and more technology dependent day by day. Nevertheless, agriculture is imperative for our survival. Rice is one of the primary food grains. It provides sustenance to almost fifty percent of the world population and promotes huge amount of employments. Hence, proper mitigation of rice plant diseases is of paramount importance. A model to detect three rice leaf diseases, namely bacterial leaf blight, brown spot, and leaf smut is proposed in this paper. Backgrounds of the images are removed by saturation threshold while disease affected areas are segmented using hue threshold. Distinctive features from color, shape, and texture domain are extracted from affected areas. These features can robustly describe local and global statistics of such images. Trying a couple of classification algorithms, extreme gradient boosting decision tree ensemble is incorporated in this model for its superior performance. Our model achieves 86.58% accuracy on rice leaf diseases dataset from UCI, which is higher than previous works on the same dataset. Class-wise accuracy of the model is also consistent among the classes.

No abstract

ABSTRACT:Land cover classification has many applications like forest management, urban planning, land use change identification and environment change analysis. The passive sensing of hyperspectral systems can be effective in describing the phenomenology of the observed area over hundreds of (narrow) spectral bands. On the other hand, the active sensing of LiDAR (Light Detection and Ranging) systems can be exploited for characterising topographical information of the area. As a result, the joint use of hyperspectral and LiDAR data provides a source of complementary information, which can greatly assist in the classification of complex classes. In this study, we fuse hyperspectral and LiDAR data for land cover classification. We do a pixel-wise classification on a disjoint set of training and testing samples for five different classes. We propose a new feature combination by fusing features from both hyperspectral and LiDAR, which achieves competent classification accuracy with low feature dimension, while the existing method requires high dimensional feature vector to achieve similar classification result. Also, for the reduction of the dimension of the feature vector, Principal Component Analysis (PCA) is used as it captures the variance of the samples with a limited number of Principal Components (PCs). We tested our classification method using PCA applied on hyperspectral bands only and combined hyperspectral and LiDAR features. Classification with support vector machine (SVM) and decision tree shows that our feature combination achieves better classification accuracy compared to the existing feature combination, while keeping the similar number of PCs. The experimental results also show that decision tree performs better than SVM and requires less execution time.

Multisource remote sensing data contain complementary information on land covers, but fusing them is a challenging problem due to the heterogeneous nature of the data. This article aims to extract and integrate information from hyperspectral image (HSI) and light detection and ranging (LiDAR) data for land cover classification. As there is a scarcity of a large number of training samples for remotely sensed hyperspectral and LiDAR data, in this article, we propose a model, which is able to perform impressively using a limited number of training samples by extracting effective features representing different characteristics of objects of interest from these two complementary data sources (HSI and LiDAR). A novel feature extraction method named inverse coefficient of variation (ICV) is introduced for HSI, which considers the Gaussian probability of neighborhood between every pair of bands. We, then, propose a two-stream feature fusion approach to integrate the ICV feature with several features extracted from HSI and LiDAR data. We incorporate a fusion unit named canonical correlation analysis as a basic unit for fusing two different sets of features within each stream. We also incorporate the concept of ensemble classification where the features produced by two-stream fusion are distributed into subsets and transformed to improve the feature quality. We compare our method with the existing state-ofthe-art methods, which are based on deep learning or handcrafted feature extraction or using both of them. Experimental results show that our proposed approach performs better than other existing methods with a limited number of training samples. Index Terms-Canonical correlation analysis (CCA), fusion, hyperspectral, light detection and ranging (LiDAR), multisource.Farah Jahan received the B.Sc.(Hons.) degree in computer science and engineering from the University of Chittagong, Chittagong, Bangladesh, in 2008, the M.E. degree in information and telecommunication engineering

Nowadays, data are the most valuable content in the world. In the age of big data, we are generating quintillions of data daily in the form of text, image, video, etc. Among them, images are highly used in daily communications. Various types of images, e.g., medical images, military images, etc. are highly confidential. But, due to data vulnerabilities, transmitting such images in a secured way is a great challenge. For this reason, researchers proposed different image cryptography algorithms. Recently, biological deoxyribonucleic acid (DNA)-based concepts are getting popular for ensuring image security as well as encryption as they show good performance. However, these DNA-based methods have some limitations, e.g., these are not dynamic and their performance results are far from ideal values. Further, these encryption methods usually involve two steps, confusion and diffusion. Confusion increases huge time complexity and needs to send one or more additional map tables with a cipher to decrypt the message. In this research, we propose a novel and efficient DNA-based key scrambling technique for image encryption that addresses the above limitations. We evaluate our proposed method using 15 different datasets and achieved superior performance scores of entropy, keyspace, cipher pixel correlations, variance of histogram, time complexity and PSNR. The experimental results show that our method can be used for image encryption with a high level of confidentiality.