Revise-Net: Exploiting Reverse Attention Mechanism for Salient Object Detection

Abstract

Recently, deep learning-based methods, especially utilizing fully convolutional neural networks, have shown extraordinary performance in salient object detection. Despite its success, the clean boundary detection of the saliency objects is still a challenging task. Most of the contemporary methods focus on exclusive edge detection modules in order to avoid noisy boundaries. In this work, we propose leveraging on the extraction of finer semantic features from multiple encoding layers and attentively re-utilize it in the generation of the final segmentation result. The proposed Revise-Net model is divided into three parts: (a) the prediction module, (b) a residual enhancement module, and (c) reverse attention modules. Firstly, we generate the coarse saliency map through the prediction modules, which are fine-tuned in the enhancement module. Finally, multiple reverse attention modules at varying scales are cascaded between the two networks to guide the prediction module by employing the intermediate segmentation maps generated at each downsampling level of the REM. Our method efficiently classifies the boundary pixels using a combination of binary cross-entropy, similarity index, and intersection over union losses at the pixel, patch, and map levels, thereby effectively segmenting the saliency objects in an image. In comparison with several state-of-the-art frameworks, our proposed Revise-Net model outperforms them with a significant margin on three publicly available datasets, DUTS-TE, ECSSD, and HKU-IS, both on regional and boundary estimation measures.

Empirical Analysis of Early Childhood Enlightenment Education Using Neural Network

“…Making full use of this semantic association can mine various network behavior characteristics of early childhood enlightenment education from different dimensions. It helps to establish a comprehensive cross-media big data knowledge association of early childhood enlightenment education and thus plays a good role in promoting the accurate search of cross-media big data [21]. e characteristics of different forms of educational data, such as text and image, are heterogeneous, and there is a large semantic gap.…”

A Novel Image Encryption Algorithm Based on Improved Arnold Transform and Chaotic Pulse-Coupled Neural Network

“…The algorithm first scrambles the rows and columns of the image and then diffuses the pixel values to obtain an encrypted image. Gopal Ghosh et al [ 8 , 9 ] proposed a security monitoring framework for IoT systems based on image encryption. The initial parameters of the hyperchaotic map are obtained based on the partially regenerated non-dominated optimization algorithm (PRNDO).…”

An Improved Deep Residual Convolutional Neural Network for Plant Leaf Disease Detection

“…The DCNNs are most successful in image classification and object detection tasks [ 7 ]. A large volume of data is required to train the DCNN models for use in various domains [ 8 ]. The data augmentation technique was introduced to increase the amount of training data without data collection for better training performance of DCNN models [ 9 ].…”