A Novel Shape Feature Descriptor for the Classification of Polyps in HD Colonoscopy

A vehicle make and model recognition (VMMR) system is a common requirement in the field of intelligent transportation systems (ITS). However, it is a challenging task because of the subtle differences between vehicle categories. In this paper, we propose a hierarchical scheme for VMMR. Specifically, the scheme consists of (1) a feature extraction framework called weighted mask hierarchical bilinear pooling (WMHBP) based on hierarchical bilinear pooling (HBP) which weakens the influence of invalid background regions by generating a weighted mask while extracting features from discriminative regions to form a more robust feature descriptor; (2) a hierarchical loss function that can learn the appearance differences between vehicle brands, and enhance vehicle recognition accuracy; (3) collection of vehicle images from the Internet and classification of images with hierarchical labels to augment data for solving the problem of insufficient data and low picture resolution and improving the model’s generalization ability and robustness. We evaluate the proposed framework for accuracy and real-time performance and the experiment results indicate a recognition accuracy of 95.1% and an FPS (frames per second) of 107 for the framework for the Stanford Cars public dataset, which demonstrates the superiority of the method and its availability for ITS.

Download Full-text

The Parametric Drawing System of Shaft Based on Feature

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.757.153 ◽

2015 ◽

Vol 757 ◽

pp. 153-157

Author(s):

Zhi Xue Sun ◽

Yang Zhang

Keyword(s):

Parametric Design ◽

Optical Axis ◽

Drive Shaft ◽

Shape Feature ◽

Image Block ◽

Stepped Shaft ◽

Drawing System ◽

Other Information ◽

Object Arx

The paper mainly studied the parametric drawing system of shaft based on feature. Firstly, the shaft was classified, and its characteristic was analyzed. Then, the special menu and image block menu of the system were designed in AutoCAD. And a friendly-interface parametric drawing system of shaft was built with geometry sculpts technology, object-oriented technology and parametric design technology, based on the platform of AutoCAD, with the development tool of Object ARX. It implemented the drawing of hardware working drawing of shaft according to the selected proportion, and improved the efficiency of the shaft design.Overview of shaftShaft is one of the main parts of machinery, depending on the axis shape; shaft can be divided into two categories: direct axis and crank shaft. According to the different load of the nature, direct axis can be divided into three types: spindle, drive shaft and the shaft. According to different shape, direct axis can be divided into two kinds: optical axis and stepped shaft. Optical axis has the following characteristics: simple shape, easy processing, less stress concentration source, optical axis is mainly used for drive shaft; Stepped shaft, in contrast to the optical axis. It is often used in the shaft. The classification of the shaft can be clearly said to use figure 1. In addition, there are some special shaft, such as the camshaft and wire soft shaft, etc.Fig. 1 Classification of the shaftOn the basis of the characteristics, the shaft information model can be considered. Usually a mechanical parts contains numerous features, shape features are the most important characteristics of these, which is the carrier of other information. We can decompose research the shape feature and think that some auxiliary features and the main features combined shape feature of axis, the main characteristics of the spatial position is adjacency relations; Auxiliary features is attached to a primary, auxiliary features can be attached to the main outline of the surface or end[1].Combining with the actual situation of this system, we analyze the features of commonly used straight shaft parts; we obtained the shape of the feature classification results as shown in table1.Table 1 Straight shaft parts classification of shape feature

Download Full-text

Using Stacked Generalization to Combine SVMs in Magnitude and Shape Feature Spaces for Classification of Hyperspectral Data

IEEE Transactions on Geoscience and Remote Sensing ◽

10.1109/tgrs.2008.2010491 ◽

2009 ◽

Vol 47 (7) ◽

pp. 2193-2205 ◽

Cited By ~ 39

Author(s):

Jin Chen ◽

Cheng Wang ◽

Runsheng Wang

Keyword(s):

Hyperspectral Data ◽

Shape Feature ◽

Stacked Generalization ◽

Feature Spaces

Download Full-text

Automated Classification of Glandular Tissue by Statistical Proximity Sampling

International Journal of Biomedical Imaging ◽

10.1155/2015/943104 ◽

2015 ◽

Vol 2015 ◽

pp. 1-11 ◽

Cited By ~ 4

Author(s):

Jimmy C. Azar ◽

Martin Simonsson ◽

Ewert Bengtsson ◽

Anders Hast

Keyword(s):

Quantitative Information ◽

Multiple Instance Learning ◽

Feature Descriptor ◽

Relative Distribution ◽

Automated Classification ◽

General Applicability ◽

Implicit Representation ◽

New Feature ◽

Tubular Formation

Due to the complexity of biological tissue and variations in staining procedures, features that are based on the explicit extraction of properties from subglandular structures in tissue images may have difficulty generalizing well over an unrestricted set of images and staining variations. We circumvent this problem by an implicit representation that is both robust and highly descriptive, especially when combined with a multiple instance learning approach to image classification. The new feature method is able to describe tissue architecture based on glandular structure. It is based on statistically representing the relative distribution of tissue components around lumen regions, while preserving spatial and quantitative information, as a basis for diagnosing and analyzing different areas within an image. We demonstrate the efficacy of the method in extracting discriminative features for obtaining high classification rates for tubular formation in both healthy and cancerous tissue, which is an important component in Gleason and tubule-based Elston grading. The proposed method may be used for glandular classification, also in other tissue types, in addition to general applicability as a region-based feature descriptor in image analysis where the image represents a bag with a certain label (or grade) and the region-based feature vectors represent instances.

Download Full-text

Classification of breast tumors as benign and malignant using textural feature descriptor

2017 IEEE International Conference on Bioinformatics and Biomedicine (BIBM) ◽

10.1109/bibm.2017.8217811 ◽

2017 ◽

Cited By ~ 3

Author(s):

Mukta Sharma ◽

Rahul Singh ◽

Mahua Bhattacharya

Keyword(s):

Breast Tumors ◽

Feature Descriptor ◽

Textural Feature

Download Full-text

Separation of Commercially Important Tuna from Other Fishes Using Feature Descriptor and Pre-trained CNN Models

European Journal of Nutrition & Food Safety ◽

10.9734/ejnfs/2021/v13i430411 ◽

2021 ◽

pp. 72-82

Author(s):

Jisha Anu Jose ◽

C. Sathish Kumar ◽

S. Sureshkumar

Keyword(s):

Fish Species ◽

Misclassification Error ◽

Polynomial Kernel ◽

Feature Descriptor ◽

System A ◽

Proposed Model ◽

Commercially Important ◽

Learning Machine ◽

Export Industries

Aims / Objectives: Identification of fish species is essential in export industries. Among the different fish species exported, tuna forms a significant portion and hence the separation of tuna from other fishes is necessary. The work aims to develop automated systems for the separation of commercially important tuna from other fishes. Methodology: The work proposes two models for the classification of commercial fishes. The first model uses conventional feature descriptors, which extract features from both spatial and frequency domain. These features are combined and are reduced by an ensemble dimension reduction method. The combined and reduced feature sets are evaluated using different classifiers. The second proposed model uses four pre-trained convolutional neural networks, VGG16, VGG19, Xception, and MobileNet, for the classification. The models are fine-tuned for the classification process. Results: Results show that for the first model, extreme learning machine classifier with Mercer wavelet kernel gives high accuracy on combined feature set while the polynomial kernel ELM provides better performance with the reduced set. For the second model, a comparison of the performance of four CNN models is done, and results indicate that VGG19 outperforms other networks in the classification task. Conclusion: Among the two proposed models, pre-trained CNN based model shows better performance than the conventional method in the separation task. Different performance measures, accuracy, precision, recall, F-score, and misclassification error are used to evaluate the system. A comparison of performance of the proposed models with the state-of-the-art systems is also reported.

Download Full-text