Farming has come a long way further than directly a system to provide food to an ever-increasing population. The seventieth population of Asian countries depends on agriculture. The identification of plant leaf diseases is typically carried out with the inspection by masters and organic tests. Both techniques are manual, costly and time-consuming. Detecting plant leaf diseases is a challenging task since it involves numerous variables such as genotype, environment, and their interplay. Accurate detection of plant leaf diseases involves a fundamental understanding of functional relativity and the collaborative factors. Deep learning helps in recognition of plant diseases and it provides earlier and timely detection of plant diseases for efficient disease management. It expresses a different infusion that gives systematic and automated detection of diseases. A convolutional neural network model is designed to use for detection of leaf spot diseases in plants. A tomato plant disease dataset has been generated from the local farms of Chittoor District, Andhra Pradesh, India. The tomato leaf spot diseases considered for training are Septoria leaf spot, bacterial spot, target spot and leaf mold which are most popular diseases in tomato plants. The average classification accuracy of the proposed model is determined to be 90.6%. For comparative analysis, the proposed CNN model has been experimentally evaluated with Multilayer perceptron classifier model. The proposed model produced an effective average MSE and R <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> values of 0.016 and 0.991. The MLP model produced less optimal results than the proposed model thus confirming the efficacy of the proposed model in classifying the plant leaf diseases. 

Modeling Convolutional Neural Network for Detection of Plant Leaf Spot Diseases

Farming has come a long way further than directly a system to provide food to an ever-increasing population. The seventieth population of Asian countries depends on agriculture. The identification of plant leaf diseases is typically carried out with the inspection by masters and organic tests. Both techniques are manual, costly and time-consuming. Detecting plant leaf diseases is a challenging task since it involves numerous variables such as genotype, environment, and their interplay. Accurate detection of plant leaf diseases involves a fundamental understanding of functional relativity and the collaborative factors. Deep learning helps in recognition of plant diseases and it provides earlier and timely detection of plant diseases for efficient disease management. It expresses a different infusion that gives systematic and automated detection of diseases. A convolutional neural network model is designed to use for detection of leaf spot diseases in plants. A tomato plant disease dataset has been generated from the local farms of Chittoor District, Andhra Pradesh, India. The tomato leaf spot diseases considered for training are Septoria leaf spot, bacterial spot, target spot and leaf mold which are most popular diseases in tomato plants. The average classification accuracy of the proposed model is determined to be 90.6%. For comparative analysis, the proposed CNN model has been experimentally evaluated with Multilayer perceptron classifier model. The proposed model produced an effective average MSE and R2 values of 0.016 and 0.991. The MLP model produced less optimal results than the proposed model thus confirming the efficacy of the proposed model in classifying the plant leaf diseases.

Change detection is an essential task to study the changes of earth surfaces in remote sensing. It is extensively being investigated in SAR images nowadays. However, SAR images suffer from the speckle noise, which is a major drawback. To address the speckle noise problem, we propose the convolutional neural network with curvelet transform. As curvelet transform can suppress the noise, it is applied in preclassification, where the difference image is transformed using curvelet. Further, transformed image is classified by hierarchical fuzzy c-means (FCM) clustering, where each pixel is classified into different classes like changed class and unchanged class. From the preclassification, patches centered at the pixels belonging to these classes are generated as the training samples. Moreover, these training samples are passed through median filter before sending them to the convolutional neural network (CNN). The median filter helps in the reduction of noise. After the training of the CNN model, the trained model classifies the image pixels and provides the final binary change map. The experimental results obtained from the two SAR datasets confirm the effectiveness of the proposed method. 

Change Detection of SAR images based on Convolution Neural Network with Curvelet Transform

Forward-looking sonar is a technique widely used for underwater detection. However, most sonar images have underwater noise and low resolution due to their acoustic properties. In recent years, the semantic segmentation model U-Net has shown excellent segmentation performance, and it has great potential in forward-looking sonar image segmentation. However, forward-looking sonar images are affected by noise, which prevents the existing U-Net model from segmenting small objects effectively. Therefore, this study presents a forward-looking sonar semantic segmentation model called Feature Pyramid U-Net with Attention (FPUA). This model uses residual blocks to improve the training depth of the network. To improve the segmentation accuracy of the network for small objects, a feature pyramid module combined with an attention structure is introduced. This improves the model’s ability to learn deep semantic and shallow detail information. First, the proposed model is compared against other deep learning models and on two datasets, of which one was collected in a tank environment and the other was collected in a real marine environment. To further test the validity of the model, a real forward-looking sonar system was devised and employed in the lake trials. The results show that the proposed model performs better than the other models for small-object and few-sample classes and that it is competitive in semantic segmentation of forward-looking sonar images.

/pdf/feature-pyramid-u-net-with-attention-for-semantic-12pvss2x.pdf

Feature Pyramid U-Net with Attention for Semantic Segmentation of Forward-Looking Sonar Images

The remote sensing (RS) field has an increasing research interest in using deep learning (DL) models to recognize kinds of RS data, leading to a great demand for training data annotation. Due to the high cost of expertise, using nonexperts to label data has become an important way to improve labeling efficiency. Commonly, a single data sample is labeled by multiple annotators and the most voted label is accepted to promise accuracy. But in the RS context, the widely admitted strategy could lose effect. Usually RS data involve considerable classes on account of the complexity of surface environments, which is prone to interclass similarity difficult to distinguish. Annotators without expertise probably make mistakes on these indistinguishable classes, thus causing error voted labels. Although classification of different characteristics in RS data has been widely documented, the nonexpert annotators are unfamiliar with these expertise, and it is difficult to force them to handle specialized labeling skills. To address the issues, this article bases multiannotator label selection on the investigation of annotators’ own ability in distinguishing similar classes of images. A quality evaluation process is designed which weights the labels from capable annotators higher than those from weak ones. By a multi-round quality evaluation algorithm, correct labels could outcompete the wrong ones even disadvantaged in numbers. Experimental results demonstrate the advance of the proposed method on the RS datasets.

Accurate Label Refinement From Multiannotator of Remote Sensing Data

The speckle noise present in synthetic‐aperture radar (SAR) images is responsible for hindering the extraction of the exact information that needs to be utilized for potential remote sensing applications. Thus the quality of SAR images needs to be enhanced by removing speckle noise in an effective manner. In this paper, A Deep Neural Network‐based Speckle Noise Removal Technique (DNN‐SNRT) is proposed that utilizes the benefits of convolution and Long Short Term Memory‐based neural networks to enhance the quality of SAR images. The proposed DNN‐SNRT uses multiple radar intensity images that are archived from the specific area of interest to facilitate the self‐learning of the intensity features derived from the image patches. The proposed DNN‐SNRT incorporates a dual neural network to remove speckle noise and flexibly estimates the thresholds and weights to achieve an effective SAR image quality improvement. The proposed DNN‐SNRT is capable of automatically updating the intensity features of SAR images during the training process. Experimental investigation of the proposed DNN‐SNRT conducted based on TerraSAR‐X images confirmed the superior enhancement of image quality over comparable recent filters. The results of the DNN‐SNRT scheme were also proved that it is able to reduce noise and preserve edges during the image quality enhancement process.

Loganathan Ganesh Babu

Papers

A deep neural network learning‐based speckle noise removal technique for enhancing the quality of synthetic‐aperture radar images