Gender classification from facial images using gray relational analysis with novel local binary pattern descriptors

2016 ◽  
Vol 11 (4) ◽  
pp. 769-776 ◽  
Author(s):  
Yılmaz Kaya ◽  
Ömer Faruk Ertuğrul
Keyword(s):  
Local Binary Pattern ◽  
Gray Relational Analysis ◽  
Gender Classification ◽  
Relational Analysis ◽  
Facial Images

MULTI-VIEW GENDER CLASSIFICATION USING MULTI-RESOLUTION LOCAL BINARY PATTERNS AND SUPPORT VECTOR MACHINES

2007 ◽  
Vol 17 (06) ◽  
pp. 479-487 ◽  
Author(s):  
HUI-CHENG LIAN ◽  
BAO-LIANG LU
Keyword(s):  
Feature Extraction ◽  
Support Vector Machines ◽  
Local Binary Pattern ◽  
Spatial Information ◽  
Local Binary Patterns ◽  
Support Vector ◽  
Gender Classification ◽  
Classification Rate ◽  
Vector Machines ◽  
Facial Images

In this paper, we present a novel method for multi-view gender classification considering both shape and texture information to represent facial images. The face area is divided into small regions from which local binary pattern (LBP) histograms are extracted and concatenated into a single vector efficiently representing a facial image. Following the idea of local binary pattern, we propose a new feature extraction approach called multi-resolution LBP, which can retain both fine and coarse local micro-patterns and spatial information of facial images. The classification tasks in this work are performed by support vector machines (SVMs). The experiments clearly show the superiority of the proposed method over both support gray faces and support Gabor faces on the CAS-PEAL face database. A higher correct classification rate of 96.56% and a higher cross validation average accuracy of 95.78% have been obtained. In addition, the simplicity of the proposed method leads to very fast feature extraction, and the regional histograms and fine-to-coarse description of facial images allow for multi-view gender classification.

Reason Analysis of Runway Incursions Based on Grey Theory

2011 ◽  
Vol 328-330 ◽  
pp. 2400-2404
Author(s):  
Zi Qi Ju
Keyword(s):  
Gray Relational Analysis ◽  
Grey Theory ◽  
Prevention Measures ◽  
Key Factors ◽  
Class Ab ◽  
Relational Analysis ◽  
Definition Of ◽  
Grey Relational ◽  
Runway Incursions ◽  
Runway Incursion

To prevent runway incursions, we should have the corresponding systematic prevent ideas. Based on the definition of runway incursions and classification of relevant criteria, it analyzed the runway incursion system, put forward the closed-loop management ideas to prevent runway incursions, and found the main contradictions of preventing runway incursions using the gray relational analysis. With the example of runway incursion dates of U.S.A, by means of Grey Relational Analysis of different severities and different factors for runway incursions, it have shown that the key factors leading to the class AB and class CD runway incursions are Vehicle/Pedestrian Deviations and Pilot Deviations respectively. Meanwhile, it proposed integrated prevention measures of runway incursions.

A MPTCP Path Selection Strategy Based on Improved Grey Relational Analysis

2013 ◽  
Vol 401-403 ◽  
pp. 1766-1771 ◽  
Author(s):  
Lan Kou ◽  
Si Rui Chen ◽  
Rui Wang
Keyword(s):  
Transmission Control Protocol ◽  
Path Selection ◽  
Transport Layer ◽  
Reference Sequence ◽  
Gray Relational Analysis ◽  
Selection Strategy ◽  
Relational Analysis ◽  
Multiple Paths ◽  
End To End ◽  
Grey Relational

Multipath Transmission Control Protocol (MPTCP), a transport layer protocol, proposed by the IETF working group in 2009, can provide multipath communication end to end. It also can improve the utilization of network resources and network transmission reliability. However, that how to select multiple paths to improve the end to end overall throughput, and how to avoid the throughput declining by the performance difference, become the focus of this study. We propose a path selection strategy based on improved gray relational analysis, and set the optimal values of the QoS parameters for the selected paths as the reference sequence. According to the value of improved grey relational degree (IGRD) which is compared with reference sequence, we select the paths with better performance, smaller difference for transmission.

Gender classification from unaligned facial images using support subspaces

2013 ◽  
Vol 221 ◽  
pp. 98-109 ◽  
Author(s):  
Wen-Sheng Chu ◽  
Chun-Rong Huang ◽  
Chu-Song Chen
Keyword(s):  
Gender Classification ◽  
Facial Images

OPTIMIZATION OF PROCESS PARAMETERS AFFECTING CUTTING FORCE, POWER CONSUMPTION AND SURFACE ROUGHNESS USING TAGUCHI-BASED GRAY RELATIONAL ANALYSIS IN TURNING AISI 1040 STEEL

2021 ◽  
Author(s):  
MAHIR AKGÜN
Keyword(s):  
Surface Roughness ◽  
Power Consumption ◽  
Cutting Force ◽  
Cutting Speed ◽  
System Optimization ◽  
Gray Relational Analysis ◽  
Depth Of Cut ◽  
Nose Radius ◽  
Significance Level ◽  
Relational Analysis

This study focuses on optimization of cutting conditions and modeling of cutting force ([Formula: see text]), power consumption ([Formula: see text]), and surface roughness ([Formula: see text]) in machining AISI 1040 steel using cutting tools with 0.4[Formula: see text]mm and 0.8[Formula: see text]mm nose radius. The turning experiments have been performed in CNC turning machining at three different cutting speeds [Formula: see text] (150, 210 and 270[Formula: see text]m/min), three different feed rates [Formula: see text] (0.12 0.18 and 0.24[Formula: see text]mm/rev), and constant depth of cut (1[Formula: see text]mm) according to Taguchi L18 orthogonal array. Kistler 9257A type dynamometer and equipment’s have been used in measuring the main cutting force ([Formula: see text]) in turning experiments. Taguchi-based gray relational analysis (GRA) was also applied to simultaneously optimize the output parameters ([Formula: see text], [Formula: see text] and [Formula: see text]). Moreover, analysis of variance (ANOVA) has been performed to determine the effect levels of the turning parameters on [Formula: see text], [Formula: see text] and [Formula: see text]. Then, the mathematical models for the output parameters ([Formula: see text], [Formula: see text] and [Formula: see text]) have been developed using linear and quadratic regression models. The analysis results indicate that the feed rate is the most important factor affecting [Formula: see text] and [Formula: see text], whereas the cutting speed is the most important factor affecting [Formula: see text]. Moreover, the validation tests indicate that the system optimization for the output parameters ([Formula: see text], [Formula: see text] and [Formula: see text]) is successfully completed with the Taguchi method at a significance level of 95%.

scholarly journals Gender classification using custom convolutional neural networks architecture

2020 ◽  
Vol 10 (6) ◽  
pp. 5758
Author(s):  
Fadhlan Hafizhelmi Kamaru Zaman
Keyword(s):  
Neural Network ◽  
Classification Accuracy ◽  
High Accuracy ◽  
Input Image ◽  
Gender Classification ◽  
Specific Task ◽  
Male And Female ◽  
Computation Efficiency ◽  
The Arts ◽  
Facial Images

Gender classification demonstrates high accuracy in many previous works. However, it does not generalize very well in unconstrained settings and environments. Furthermore, many proposed Convolutional Neural Network (CNN) based solutions vary significantly in their characteristics and architectures, which calls for optimal CNN architecture for this specific task. In this work, a hand-crafted, custom CNN architecture is proposed to distinguish between male and female facial images. This custom CNN requires smaller input image resolutions and significantly fewer trainable parameters than some popular state-of-the-arts such as GoogleNet and AlexNet. It also employs batch normalization layers which results in better computation efficiency. Based on experiments using publicly available datasets such as LFW, CelebA and IMDB-WIKI datasets, the proposed custom CNN delivered the fastest inference time in all tests, where it needs only 0.92ms to classify 1200 images on GPU, 1.79ms on CPU, and 2.51ms on VPU. The custom CNN also delivers performance on-par with state-of-the-arts and even surpassed these methods in CelebA gender classification where it delivered the best result at 96% accuracy. Moreover, in a more challenging cross-dataset inference, custom CNN trained using CelebA dataset gives the best gender classification accuracy for tests on IMDB and WIKI datasets at 97% and 96% accuracy respectively.

Sign in / Sign up

Export Citation Format

Share Document