Results 101 to 110 of about 4,627,276 (412)
Force field feature extraction for ear biometrics
The overall objective in defining feature space is to reduce the dimensionality of the original pattern space, whilst maintaining discriminatory power for classification.
Carter, John N.+2 more
core +1 more source
Brain cancer classification is an important step that depends on the physician’s knowledge and experience. An automated tumor classification system is very essential to support radiologists and physicians to identify brain tumors.
Abdu H. Gumaei+4 more
semanticscholar +1 more source
The mitochondrial outer membrane iron–sulphur ([Fe‐S]) protein mitoNEET is a target of the type‐2 diabetes drug pioglitazone. Its unknown molecular function is linked to respiratory complex activity and mitochondrial function. We discovered that O2 protects the mitoNEET [2Fe‐2S] cluster against NO oxidation and desensitization towards reduction by H2S.
Thao Nghi Hoang+9 more
wiley +1 more source
Application of Improved DNN Algorithm Based on Feature Fusion in Fine-Grained Image Recognition
Fine-grained image recognition is a research highlight in the computer vision. Compared with traditional image classification tasks, it places more emphasis on distinguishing objects with similar appearance features but belonging to different categories.
Jiongguang Zhu, Wei Zhang
doaj +1 more source
A Novel Scalable Apache Spark Based Feature Extraction Approaches for Huge Protein Sequence and their Clustering Performance Analysis [PDF]
Genome sequencing projects are rapidly increasing the number of high-dimensional protein sequence datasets. Clustering a high-dimensional protein sequence dataset using traditional machine learning approaches poses many challenges. Many different feature extraction methods exist and are widely used. However, extracting features from millions of protein
arxiv
Characterization of fungal carbonyl sulfide hydrolase belonging to clade D β‐carbonic anhydrase
Here, we performed a functional analysis of the fungal enzymes belonging to clade D of the β‐class carbonic anhydrase family (β‐D‐CA). The β‐D‐CAs in the basidiomycete Gloeophyllum trabeum and the ascomycete Trichoderma harzianum showed very low activity in the hydration of CO2 but exhibited high activity in the hydrolysis of carbonyl sulfide (COS ...
Ryuka Iizuka+6 more
wiley +1 more source
Global Regular Network for Writer Identification [PDF]
Writer identification has practical applications for forgery detection and forensic science. Most models based on deep neural networks extract features from character image or sub-regions in character image, which ignoring features contained in page-region image. Our proposed global regular network (GRN) pays attention to these features.
arxiv
Texture Feature Extraction by Using Local Binary Pattern [PDF]
Local Binary Pattern (LBP) is a method that used to describe texture characteristics of the surfaces. By applying LBP, texture pattern probability can be summarised into a histogram. LBP values need to be determined for all of the image pixels.
Prakasa, E. (Esa)
core
In modular polyketide synthases, the dehydratase (DH) domain catalyzes the dehydration reaction of the β‐hydroxyacyl unit attached to the cognate acyl carrier protein (ACP) domain. However, it is unclear how DH interacts with ACP during the reaction. In this study, we identified DH–ACP interface residues, providing the first detailed insights into DH ...
Kaede Kotagiri+8 more
wiley +1 more source
Convolution channel separation and frequency sub-bands aggregation for music genre classification [PDF]
In music, short-term features such as pitch and tempo constitute long-term semantic features such as melody and narrative. A music genre classification (MGC) system should be able to analyze these features. In this research, we propose a novel framework that can extract and aggregate both short- and long-term features hierarchically.
arxiv