Divya Mishra

Divya Mishra

Auto-generated profile, based on publicly available information

Divya Mishra

School of Electrical and Computer Engineering, Ben Gurion University of the Negev, Beersheba, Israel

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
25 January 2023 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Verified
Journal Article
Open access
DMDivya Mishra
Divya Mishra
OHOfer Hadar
Ofer Hadar
Real-world degradations deviate from ideal degradations, as most deep learning-based scenarios involve the ideal synthesis of low-resolution counterpart images by popularly used bicubic interpolation. Moreover, supervised learning approaches rely on many high-resolution and low-resolution image pairings to reconstruct missing information based on their association, developed by complex long hours of deep neural network training. Additionally, the trained model's generalizability on various image datasets with various distributions is not guaranteed. To overcome this challenge, we proposed our novel Self-FuseNet, particularly for extremely poor-resolution satellite images. Also, the network exhibits strong generalization performance on additional datasets (both “Ideal” and “Non-Ideal” scenarios). The network is especially for those image datasets suffering from two significant limitations: (1) non-availability of ground truth high-resolution images and (2) limitation of a large count of the unpaired dataset for deep neural network training. The benefit of the proposed model is threefold: First, it does not require any significant extensive training data, either paired or unpaired but only a single low-resolution image without prior knowledge of its distribution. Secondly, it is a simple and effective model for super-resolving very poor-resolution images, saving computational resources and time. Third, using UNet, the processing of data is accelerated by the network's wide skip connections, allows image reconstruction with fewer parameters. Rather than using an inverse approach, as common in most deep learning scenarios, we introduced a forward approach to super-resolve exceptionally low-resolution remote sensing images. This demonstrates its supremacy over recently proposed state-of-the-art methods for unsupervised single real-world image blind super-resolution.

12 July 2023 in IEEE Geoscience and Remote Sensing Letters
Verified
Journal Article
Closed access
DMDivya Mishra
Divya Mishra
OHOfer Hadar
Ofer Hadar
Real-world degradations diverge from ideal degradations since most self-supervised and unsupervised learning scenarios generate low-resolution fake counterpart images from existing data using a common bicubic kernel. Additionally, conventional unsupervised learning techniques rely on a large number of training samples with excessive diversity as an inevitable requirement to reconstruct missing data based on their downsampled correlation. Practically, it is time-consuming to arrange large counts of samples along with the diversity for training. In this letter, we proposed a network CLSR: Contrastive Learning for remote sensing Image Super-Resolution in a semi-supervised setting. Contrastive learning is the idea of comparing two samples to find shared features and attributes that set one data class apart, thus boosting visual task performance. Experiments demonstrate that it can super-resolve different modalities of data: single-band, multi-spectral band, RGB remote sensing images, and real-world natural images.

11 September 2023 in IEEE Transactions on Geoscience and Remote Sensing
Verified
Journal Article
Closed access
DMDivya Mishra
Divya Mishra
OHOfer Hadar
Ofer Hadar
Contrastive learning is a self-supervised comparison of two samples to identify characteristics and traits that distinguish one data class from another, improving performance on visual tasks. The performance of existing super-resolution-based approaches degrades with increasing scaling factors, hence practically not useful for high-resolution imaging applications. We proposed a novel framework that uses contrastive training followed by a decoder to generate an "Artificial style image," which is utilized as a style image for neural style transfer learning for image super-resolution in an unsupervised manner. The idea is to benefit from high-resolution textures and features as a style and transfer on an original low-resolution content image as base elements. The proposed framework has three benefits: (1) The framework is capable of super-resolving different modalities of data like single-band remote sensing images, multi-spectral band images, RGB remote sensing images, and real-world natural images. (2) Proposed method outperforms existing unsupervised and also supervised learning-based methods for both visual and qualitative performance. (3) Leveraging neural style-transfer learning for remote sensing image super-resolution is performed without sacrificing speed and resources. The framework is novel since the work on neural style transfer learning to super-resolve remote sensing images in an unsupervised manner has yet to be acknowledged.

6 February 2024 in Multimedia Tools and Applications
Verified
Journal Article
Open access
PDF file
RSRon Shmueli
Ron Shmueli
DMDivya Mishra
Divya Mishra
TSTal Shmueli
Tal Shmueli
Ofer HadarOfer Hadar
Ofer Hadar
Image steganography is the art of hiding information in a cover image in such a way that a third party does not notice the hidden information. This paper presents a novel technique for image steganography in the spatial domain. The new method hides and recovers hidden information of substantial length within digital imagery while maintaining the size and quality of the original image. The image gradient is used to generate a saliency image, which represents the energy of each pixel in the image. Pixels with higher energy are more salient and they are valuable for hiding data since their visual impairment is low. From the saliency image, a cumulative maximum energy matrix is created; this matrix is used to generate horizontal seams that pass over the maximum energy path. By embedding the secret bits of information along the seams, a stego-image is created which contains the hidden message. In the stego-image, we ensure that the hidden data is invisible, with very small perceived image quality degradation. The same algorithms are used to reconstruct the hidden message from the stego-image. Experiments have been conducted using two types of images and two types of hidden data to evaluate the proposed technique. The experimental results show that the proposed algorithm has a high capacity and good invisibility, with a Peak Signal-to-Noise Ratio (PSNR) of about 70, and a Structural SIMilarity index (SSIM) of about 1.

26 February 2024 in IEEE Transactions on AgriFood Electronics
Verified
Journal Article
Closed access
DMDivya Mishra
Divya Mishra
SCSharon Chemweno
Sharon Chemweno
Ofer HadarOfer Hadar
Ofer Hadar
OBOfer Ben-Tovim
Ofer Ben-Tovim
NLNaftali Lazarovitch
Naftali Lazarovitch
JEJhonathan Ephrath
Jhonathan Ephrath
Root hairs are essential for nutrient uptake and plant-microbe interactions, playing a vital role in plant health and agricultural productivity. They extend from the surface of root cells, significantly increasing the root surface area, and constitute roughly 70% of the total root area. Given these advantages, detecting root hairs in scenes with low resolution is challenging. Therefore, we have proposed a study that utilizes unsupervised image super-resolution methods to reconstruct finer details for root hairs using the dataset captured from our novel scanning camera known as RootCam. RootCam is a fully automated tool designed for monitoring and capturing plant root images for different vision tasks for a more accurate representation of their morphology and root trait measurements. Root hair super-resolution proves to be a powerful tool for root biology and its applications in precision agriculture. To the best of the authors' knowledge, this research study is the first that mainly focuses on root hairs and their trait measurement improvement using super-resolution. By visualizing the rhizosphere in high-resolution detail, we are able to notice a significant improvement in bell-pepper plant root hair count from 7 to 12, total root length from 0.32 to 1 mm, and root hair density (number of root hairs/mm) from 2.7 to 4.63, as upscaling factors rise from 2 to 8, respectively, when compared with bicubic and contrastive learning semi-supervised remote sensing image super (CLSR) for super-resolution. Researchers and farmers can make informed decisions about nutrient placement, irrigation management, and crop selection, optimizing resource use efficiency and crop yields.

21 June 2023
Verified
Book Chapter
Closed access
PDF file
DMDivya Mishra
Divya Mishra
IDItai Dror
Itai Dror
Ofer HadarOfer Hadar
Ofer Hadar
DCDaniel Choukroun
Daniel Choukroun
SMShimrit Maman
Shimrit Maman
DBDan G Blumberg
Dan G Blumberg