Abstract There is a need for automatic systems that can reliably detect, track and classify fish other marine species in underwater videos without human intervention. Conventional computer vision techniques do not perform well conditions where the background complex shape textural features of are subtle. Data-driven classification models like neural networks require huge amount labelled data, otherwise they tend to over-fit training data fail on unseen test which involved training. We...

With the advancement of powerful image processing and machine learning techniques, CAD has become ever more prevalent in all fields medicine including ophthalmology. Since optic disc is most important part retinal fundus for glaucoma detection, this paper proposes a two-stage framework that first detects localizes then classifies it into healthy or glaucomatous. The stage based on RCNN responsible localizing extracting from while second uses Deep CNN to classify extracted In addition...

The Netherlands Forensic Institute and the for Science in Shanghai are search of a signature verification system that can be implemented forensic casework research to objectify results. We want bridge gap between recent technological developments casework. In collaboration with German Research Center Artificial Intelligence we have organized competition on datasets two scripts (Dutch Chinese) which asked compare questioned signatures against set reference signatures. received 12 systems from...

This paper presents a novel approach for the detection of tables present in documents, leveraging potential deep neural networks. Conventional approaches table rely on heuristics that are error prone and specific to dataset. In contrast, presented harvests data recognize arbitrary layout. Most prior only applicable PDFs, whereas, directly works images making it generally any format. The is based combination deformable CNN with faster R-CNN/FPN. has fixed receptive field which problematic...

Propensity of skin diseases to manifest in a variety forms, lack and maldistribution qualified dermatologists, exigency timely accurate diagnosis call for automated Computer-Aided Diagnosis (CAD). This study aims at extending previous works on CAD dermatology by exploring the potential Deep Learning classify hundreds diseases, improving classification performance, utilizing disease taxonomy. We trained state-of-the-art Neural Networks two largest publicly available image datasets, namely...

This paper presents a deep Convolutional Neural Network (CNN) based approach for document image classification. One of the main requirement CNN architecture is that they need huge number samples training. To overcome this problem we adopt which trained using big dataset containing millions i.e., ImageNet. The proposed work outperforms both traditional structure similarity methods and approaches earlier. accuracy with merely 20 images per class state-of-the-art by achieving classification...

This paper presents the results of ICDAR2013 competitions on signature verification and writer identification for on- offline skilled forgeries jointly organized by PR researchers Forensic Handwriting Examiners (FHEs). The aim is to bridge gap between recent technological developments forensic casework. Two modalities (signatures, handwritten text) are considered where training evaluation data (in Dutch Japanese) were collected provided FHEs PR-researchers. Four tasks defined systems had...

Scarcity of large publicly available retinal fundus image datasets for automated glaucoma detection has been the bottleneck successful application artificial intelligence towards practical Computer-Aided Diagnosis (CAD). A few small that are research community usually suffer from impractical capturing conditions and stringent inclusion criteria. These shortcomings in already limited choice existing make it challenging to mature a CAD system so can perform real-world environment. In this...

This paper presents the results of ICDAR 2015 competition on signature verification and writer identification for on- off-line skilled forgeries jointly organized by PR-researchers Forensic Handwriting Examiners (FHEs). The aim is to bridge gap between recent technological developments forensic casework. Two modalities (signatures handwritten text) are considered training evaluation data collected provided FHEs PR-researchers. Four tasks defined four different languages; Bengali...

Eisenbergiella tayi is a pathogen that affects the oral cavity, gastrointestinal tract, skin, and vagina, shows some resistance to existing antibiotics. Identifying new antibiotic targets through computational methods could expedite process. At same time, there are numerous opportunities develop antibiotics address infections caused by this pathogen. In study, proteome of E. was progressively reduced pinpoint potential targets. The main goals were identify proteins non-redundant, unique...

E. coli strains have been isolated and identified in Pakistan, but detailed genomic data specific to these is still lacking. This gap knowledge prevents a full understanding of coli’s genomics the region. Research indicates that both phagocytic nonphagocytic mammalian cells can use Shigella, Listeria, Salmonella invasive as vectors for gene delivery. study seeks investigate epidemiology pathogenic features Pakistan. Stool samples were collected from various regions exhibiting severe symptoms...

This study examines the economic and geopolitical consequences of President Donald Trump’s “America First” policy on EU-US trade relations between 2017 2021. Through a realist theoretical lens, paper analyzes three key case studies—steel aluminum tariffs, collapse Transatlantic Trade Investment Partnership (TTIP), digital services tax dispute—to assess how unilateral protectionism disrupted transatlantic dynamics. Findings reveal that policies eroded trust in multilateral institutions,...

This paper presents a novel two-stream approach for document image classification. The proposed leverages textual and visual modalities to classify images into ten categories, including letter, memo, news article, etc. In order alleviate dependency of stream on performance underlying OCR (which is the case with general content based classifiers), we utilize filter feature-ranking algorithm. algorithm ranks features each class their ability discriminate selects set top 'K' that are retained...

Deep learning based medical image classifiers have shown remarkable prowess in various application areas like ophthalmology, dermatology, pathology, and radiology. However, the acceptance of these Computer-Aided Diagnosis (CAD) systems real clinical setups is severely limited primarily because their decision-making process remains largely obscure. This work aims at elucidating a deep classifier by verifying that model learns utilizes similar disease-related concepts as described employed...

This paper presents a Convolutional Neural Networks (CNN) based Unsupervised Curriculum Learning approach for the recognition of water bodies to overcome stated challenges remote sensing RGB imagery. The unsupervised nature presented algorithm eliminates need labelled training data. problem is cast as two class clustering (water and non-water), while done on deep features obtained by pre-trained CNN. After initial clusters have been identified, representative samples from each cluster are...

One of the major reasons death worldwide is Cardiovascular diseases. its type Arrhythmia in which normal rhythm heart varied due to damage muscles and electrolyte imbalance. To study cardiovascular disease, Electrocardiogram (ECG) signal used that plays a significant role identifying Arrhythmia. By using combination Multivariate Empirical Mode Decomposition (MEMD) Artificial Neural Network (ANN), hybrid technique proposed this research work detect classify MEMD, promising nowadays for...

Protein-protein interaction (PPI) prediction is essential to understand the functions of proteins in various biological processes and their roles development, progression, treatment different diseases. To perform economical large-scale PPI analysis, several artificial intelligence-based approaches have been proposed. However, these limited predictive performance due use in-effective statistical representation learning methods predictors that lack ability extract comprehensive discriminative...