Learning Attention-Aware Interactive Features for Fine-Grained Vegetable and Fruit Classification

Vegetable and fruit recognition can be considered as a fine-grained visual categorization (FGVC) task, which is challenging due to the large intraclass variances and small interclass variances. A mainstream direction to address the challenge is to exploit fine-grained local/global features to enhance the feature extraction and representation in the learning pipeline. However, unlike the human visual system, most of the existing FGVC methods only extract features from individual images during training. In contrast, human beings can learn discriminative features by comparing two different images. Inspired by this intuition, a recent FGVC method, named Attentive Pairwise Interaction Network (API-Net), takes as input an image pair for pairwise feature interaction and demonstrates superior performance in several open FGVC data sets. However, the accuracy of API-Net on VegFru, a domain-specific FGVC data set, is lower than expected, potentially due to the lack of spatialwise attention. Following this direction, we propose an FGVC framework named Attention-aware Interactive Features Network (AIF-Net) that refines the API-Net by integrating an attentive feature extractor into the backbone network. Specifically, we employ a region proposal network (RPN) to generate a collection of informative regions and apply a biattention module to learn global and local attentive feature maps, which are fused and fed into an interactive feature learning subnetwork. The novel neural structure is verified through extensive experiments and shows consistent performance improvement in comparison with the SOTA on the VegFru data set, demonstrating its superiority in fine-grained vegetable and fruit recognition. We also discover that a concatenation fusion operation applied in the feature extractor, along with three top-scoring regions suggested by an RPN, can effectively boost the performance.

Download Full-text

MSST-Net: A Multi-Scale Adaptive Network for Building Extraction from Remote Sensing Images Based on Swin Transformer

Remote Sensing ◽

10.3390/rs13234743 ◽

2021 ◽

Vol 13 (23) ◽

pp. 4743

Author(s):

Wei Yuan ◽

Wenbo Xu

Keyword(s):

Neural Network ◽

Remote Sensing ◽

Convolutional Neural Network ◽

Network Model ◽

Remote Sensing Images ◽

Feature Maps ◽

Global Features ◽

Adaptive Network ◽

Data Set ◽

Multi Scale

The segmentation of remote sensing images by deep learning technology is the main method for remote sensing image interpretation. However, the segmentation model based on a convolutional neural network cannot capture the global features very well. A transformer, whose self-attention mechanism can supply each pixel with a global feature, makes up for the deficiency of the convolutional neural network. Therefore, a multi-scale adaptive segmentation network model (MSST-Net) based on a Swin Transformer is proposed in this paper. Firstly, a Swin Transformer is used as the backbone to encode the input image. Then, the feature maps of different levels are decoded separately. Thirdly, the convolution is used for fusion, so that the network can automatically learn the weight of the decoding results of each level. Finally, we adjust the channels to obtain the final prediction map by using the convolution with a kernel of 1 × 1. By comparing this with other segmentation network models on a WHU building data set, the evaluation metrics, mIoU, F1-score and accuracy are all improved. The network model proposed in this paper is a multi-scale adaptive network model that pays more attention to the global features for remote sensing segmentation.

Download Full-text

FINET: Fine-grained Feature Interaction Network for Click-through Rate Prediction

2020 12th International Conference on Advanced Computational Intelligence (ICACI) ◽

10.1109/icaci49185.2020.9177810 ◽

2020 ◽

Author(s):

Zhibin Lian ◽

Hong Ge

Keyword(s):

Interaction Network ◽

Feature Interaction ◽

Fine Grained ◽

Rate Prediction ◽

Click Through Rate

Download Full-text

Ambiguity Reduction through Optimal Set of Region Selection Using GA and BFO for Handwritten Bangla Character Recognition

Advances in Computational Intelligence and Robotics - Handbook of Research on Swarm Intelligence in Engineering ◽

10.4018/978-1-4666-8291-7.ch019 ◽

2015 ◽

pp. 611-639

Author(s):

Nibaran Das ◽

Subhadip Basu ◽

Mahantapas Kundu ◽

Mita Nasipuri

Keyword(s):

Character Recognition ◽

Recognition System ◽

Human Beings ◽

Bacterial Foraging Optimization ◽

Global Features ◽

Data Set ◽

Region Selection ◽

Longest Run ◽

Ambiguity Reduction ◽

Optimal Set

To recognize different patterns, identification of local regions where the pattern classes differ significantly is an inherent ability of the human cognitive system. This inherent ability of human beings may be imitated in any pattern recognition system by incorporating the ability of locating the regions that contain the maximum discriminating information among the pattern classes. In this chapter, the concept of Genetic Algorithm (GA) and Bacterial Foraging Optimization (BFO) are discussed to identify those regions having maximum discriminating information. The discussion includes the evaluation of the methods on the sample images of handwritten Bangla digit and Basic character, which is a subset of Bangla character set. Different methods of sub-image or local region creation such as random creation or based on the Center of Gravity (CG) of the foreground pixels are also discussed here. Longest run features, extracted from the generated local regions, are used as local feature in the present chapter. Based on these extracted local features, together with global features, the algorithms are applied to search for the optimal set of local regions. The obtained results are higher than that results obtained without optimization on the same data set.

Download Full-text

Ambiguity Reduction Through Optimal Set of Region Selection Using GA and BFO for Handwritten Bangla Character Recognition

Computer Vision ◽

10.4018/978-1-5225-5204-8.ch053 ◽

2018 ◽

pp. 1279-1306

Author(s):

Nibaran Das ◽

Subhadip Basu ◽

Mahantapas Kundu ◽

Mita Nasipuri

Keyword(s):

Character Recognition ◽

Recognition System ◽

Human Beings ◽

Bacterial Foraging Optimization ◽

Global Features ◽

Data Set ◽

Region Selection ◽

Longest Run ◽

Ambiguity Reduction ◽

Optimal Set

Download Full-text

Object–Part Registration–Fusion Net for Fine-Grained Image Classification

Symmetry ◽

10.3390/sym13101838 ◽

2021 ◽

Vol 13 (10) ◽

pp. 1838

Author(s):

Chih-Wei Lin ◽

Mengxiang Lin ◽

Jinfu Liu

Keyword(s):

Feature Fusion ◽

Bird Species ◽

Image Understanding ◽

Local Features ◽

Feature Maps ◽

Global Features ◽

Fine Grained ◽

Object Part ◽

Global And Local ◽

Fusion Feature

Classifying fine-grained categories (e.g., bird species, car, and aircraft types) is a crucial problem in image understanding and is difficult due to intra-class and inter-class variance. Most of the existing fine-grained approaches individually utilize various parts and local information of objects to improve the classification accuracy but neglect the mechanism of the feature fusion between the object (global) and object’s parts (local) to reinforce fine-grained features. In this paper, we present a novel framework, namely object–part registration–fusion Net (OR-Net), which considers the mechanism of registration and fusion between an object (global) and its parts’ (local) features for fine-grained classification. Our model learns the fine-grained features from the object of global and local regions and fuses these features with the registration mechanism to reinforce each region’s characteristics in the feature maps. Precisely, OR-Net consists of: (1) a multi-stream feature extraction net, which generates features with global and various local regions of objects; (2) a registration–fusion feature module calculates the dimension and location relationships between global (object) regions and local (parts) regions to generate the registration information and fuses the local features into the global features with registration information to generate the fine-grained feature. Experiments execute symmetric GPU devices with symmetric mini-batch to verify that OR-Net surpasses the state-of-the-art approaches on CUB-200-2011 (Birds), Stanford-Cars, and Stanford-Aircraft datasets.

Download Full-text

A Self-Spatial Adaptive Weighting Based U-Net for Image Segmentation

Electronics ◽

10.3390/electronics10030348 ◽

2021 ◽

Vol 10 (3) ◽

pp. 348

Author(s):

Choongsang Cho ◽

Young Han Lee ◽

Jongyoul Park ◽

Sangkeun Lee

Keyword(s):

Image Segmentation ◽

Medical Image ◽

Medical Image Segmentation ◽

Feature Maps ◽

Data Set ◽

Feature Map ◽

Adaptive Weighting ◽

Spatially Adaptive ◽

Wide Range ◽

Decoder Architecture

Semantic image segmentation has a wide range of applications. When it comes to medical image segmentation, its accuracy is even more important than those of other areas because the performance gives useful information directly applicable to disease diagnosis, surgical planning, and history monitoring. The state-of-the-art models in medical image segmentation are variants of encoder-decoder architecture, which is called U-Net. To effectively reflect the spatial features in feature maps in encoder-decoder architecture, we propose a spatially adaptive weighting scheme for medical image segmentation. Specifically, the spatial feature is estimated from the feature maps, and the learned weighting parameters are obtained from the computed map, since segmentation results are predicted from the feature map through a convolutional layer. Especially in the proposed networks, the convolutional block for extracting the feature map is replaced with the widely used convolutional frameworks: VGG, ResNet, and Bottleneck Resent structures. In addition, a bilinear up-sampling method replaces the up-convolutional layer to increase the resolution of the feature map. For the performance evaluation of the proposed architecture, we used three data sets covering different medical imaging modalities. Experimental results show that the network with the proposed self-spatial adaptive weighting block based on the ResNet framework gave the highest IoU and DICE scores in the three tasks compared to other methods. In particular, the segmentation network combining the proposed self-spatially adaptive block and ResNet framework recorded the highest 3.01% and 2.89% improvements in IoU and DICE scores, respectively, in the Nerve data set. Therefore, we believe that the proposed scheme can be a useful tool for image segmentation tasks based on the encoder-decoder architecture.

Download Full-text

Properties and methods of estimation for a bivariate exponentiated Fréchet distribution

Mathematica Slovaca ◽

10.1515/ms-2017-0426 ◽

2020 ◽

Vol 70 (5) ◽

pp. 1211-1230

Author(s):

Abdus Saboor ◽

Hassan S. Bakouch ◽

Fernando A. Moala ◽

Sheraz Hussain

Keyword(s):

Maximum Likelihood ◽

Probability Density Function ◽

Probability Density ◽

Density Function ◽

Superior Performance ◽

Estimation Methods ◽

Conditional Probability Density ◽

Data Set ◽

Fréchet Distribution ◽

Frechet Distribution

AbstractIn this paper, a bivariate extension of exponentiated Fréchet distribution is introduced, namely a bivariate exponentiated Fréchet (BvEF) distribution whose marginals are univariate exponentiated Fréchet distribution. Several properties of the proposed distribution are discussed, such as the joint survival function, joint probability density function, marginal probability density function, conditional probability density function, moments, marginal and bivariate moment generating functions. Moreover, the proposed distribution is obtained by the Marshall-Olkin survival copula. Estimation of the parameters is investigated by the maximum likelihood with the observed information matrix. In addition to the maximum likelihood estimation method, we consider the Bayesian inference and least square estimation and compare these three methodologies for the BvEF. A simulation study is carried out to compare the performance of the estimators by the presented estimation methods. The proposed bivariate distribution with other related bivariate distributions are fitted to a real-life paired data set. It is shown that, the BvEF distribution has a superior performance among the compared distributions using several tests of goodness–of–fit.

Download Full-text

Ischemic Stroke Prediction by Exploring Sleep Related Features

Applied Sciences ◽

10.3390/app11052083 ◽

2021 ◽

Vol 11 (5) ◽

pp. 2083

Author(s):

Jia Xie ◽

Zhu Wang ◽

Zhiwen Yu ◽

Bin Guo ◽

Xingshe Zhou

Keyword(s):

Ischemic Stroke ◽

Life Quality ◽

Area Under The Curve ◽

Prediction Method ◽

Neural System ◽

Human Beings ◽

Filter Model ◽

Fine Grained ◽

Proposed Model ◽

Precision Recall Curve

Ischemic stroke is one of the typical chronic diseases caused by the degeneration of the neural system, which usually leads to great damages to human beings and reduces life quality significantly. Thereby, it is crucial to extract useful predictors from physiological signals, and further diagnose or predict ischemic stroke when there are no apparent symptoms. Specifically, in this study, we put forward a novel prediction method by exploring sleep related features. First, to characterize the pattern of ischemic stroke accurately, we extract a set of effective features from several aspects, including clinical features, fine-grained sleep structure-related features and electroencephalogram-related features. Second, a two-step prediction model is designed, which combines commonly used classifiers and a data filter model together to optimize the prediction result. We evaluate the framework using a real polysomnogram dataset that contains 20 stroke patients and 159 healthy individuals. Experimental results demonstrate that the proposed model can predict stroke events effectively, and the Precision, Recall, Precision Recall Curve and Area Under the Curve are 63%, 85%, 0.773 and 0.919, respectively.

Download Full-text

CADD-Splice—improving genome-wide variant effect prediction using deep learning-derived splice scores

Genome Medicine ◽

10.1186/s13073-021-00835-9 ◽

2021 ◽

Vol 13 (1) ◽

Author(s):

Philipp Rentzsch ◽

Max Schubach ◽

Jay Shendure ◽

Martin Kircher

Keyword(s):

Prediction Models ◽

Splice Variants ◽

Superior Performance ◽

Data Set ◽

Pathogenic Variants ◽

Genome Wide ◽

Donor And Acceptor ◽

Human Proteins ◽

Variant Effect ◽

Variant Effect Prediction

Abstract Background Splicing of genomic exons into mRNAs is a critical prerequisite for the accurate synthesis of human proteins. Genetic variants impacting splicing underlie a substantial proportion of genetic disease, but are challenging to identify beyond those occurring at donor and acceptor dinucleotides. To address this, various methods aim to predict variant effects on splicing. Recently, deep neural networks (DNNs) have been shown to achieve better results in predicting splice variants than other strategies. Methods It has been unclear how best to integrate such process-specific scores into genome-wide variant effect predictors. Here, we use a recently published experimental data set to compare several machine learning methods that score variant effects on splicing. We integrate the best of those approaches into general variant effect prediction models and observe the effect on classification of known pathogenic variants. Results We integrate two specialized splicing scores into CADD (Combined Annotation Dependent Depletion; cadd.gs.washington.edu), a widely used tool for genome-wide variant effect prediction that we previously developed to weight and integrate diverse collections of genomic annotations. With this new model, CADD-Splice, we show that inclusion of splicing DNN effect scores substantially improves predictions across multiple variant categories, without compromising overall performance. Conclusions While splice effect scores show superior performance on splice variants, specialized predictors cannot compete with other variant scores in general variant interpretation, as the latter account for nonsense and missense effects that do not alter splicing. Although only shown here for splice scores, we believe that the applied approach will generalize to other specific molecular processes, providing a path for the further improvement of genome-wide variant effect prediction.

Download Full-text

Knowledge Enhanced LSTM for Coreference Resolution on Biomedical Texts

Bioinformatics ◽

10.1093/bioinformatics/btab153 ◽

2021 ◽

Author(s):

Yufei Li ◽

Xiaoyong Ma ◽

Xiangyu Zhou ◽

Pengzhen Cheng ◽

Kai He ◽

...

Keyword(s):

Information Integration ◽

Short Term Memory ◽

Superior Performance ◽

Supplementary Information ◽

Specific Information ◽

Coreference Resolution ◽

Fine Grained ◽

Domain Specific ◽

Memory Network ◽

Biomedical Texts

Abstract Motivation Bio-entity Coreference Resolution focuses on identifying the coreferential links in biomedical texts, which is crucial to complete bio-events’ attributes and interconnect events into bio-networks. Previously, as one of the most powerful tools, deep neural network-based general domain systems are applied to the biomedical domain with domain-specific information integration. However, such methods may raise much noise due to its insufficiency of combining context and complex domain-specific information. Results In this paper, we explore how to leverage the external knowledge base in a fine-grained way to better resolve coreference by introducing a knowledge-enhanced Long Short Term Memory network (LSTM), which is more flexible to encode the knowledge information inside the LSTM. Moreover, we further propose a knowledge attention module to extract informative knowledge effectively based on contexts. The experimental results on the BioNLP and CRAFT datasets achieve state-of-the-art performance, with a gain of 7.5 F1 on BioNLP and 10.6 F1 on CRAFT. Additional experiments also demonstrate superior performance on the cross-sentence coreferences. Supplementary information Supplementary data are available at Bioinformatics online.

Download Full-text