Action Classification for Partially Occluded Silhouettes by Means of Shape and Action Descriptors

This paper presents an action recognition approach based on shape and action descriptors that is aimed at the classification of physical exercises under partial occlusion. Regular physical activity in adults can be seen as a form of non-communicable diseases prevention, and may be aided by digital solutions that encourages individuals to increase their activity level. The application scenario includes workouts in front of the camera, where either the lower or upper part of the camera’s field of view is occluded. The proposed approach uses various features extracted from sequences of binary silhouettes, namely centroid trajectory, shape descriptors based on the Minimum Bounding Rectangle, action representation based on the Fourier transform and leave-one-out cross-validation for classification. Several experiments combining various parameters and shape features are performed. Despite the presence of occlusion, it was possible to obtain about 90% accuracy for several action classes, with the use of elongation values observed over time and centroid trajectory.

Download Full-text

IRAC: Insecticide Resistance and Mode-of-action Classification of Insecticides

Modern Crop Protection Compounds ◽

10.1002/9783527699261.ch28 ◽

2019 ◽

pp. 995-1012 ◽

Cited By ~ 8

Author(s):

Ralf Nauen ◽

Russell Slater ◽

Thomas C. Sparks ◽

Alfred Elbert ◽

Alan Mccaffery

Keyword(s):

Insecticide Resistance ◽

Mode Of Action ◽

Action Classification

Download Full-text

CLASSIFICATION OF HIGH-DIMENSIONAL MICROARRAY DATA WITH A TWO-STEP PROCEDURE VIA A WILCOXON CRITERION AND MULTILAYER PERCEPTRON

International Journal of Computational Intelligence and Applications ◽

10.1142/s1469026811002969 ◽

2011 ◽

Vol 10 (01) ◽

pp. 1-14

Author(s):

VLADIMIR NIKULIN ◽

TIAN-HSIANG HUANG ◽

GEOFFREY J. MCLACHLAN

Keyword(s):

Data Mining ◽

Feature Selection ◽

High Dimensional ◽

Second Step ◽

Support Vector ◽

Step Procedure ◽

Leave One Out ◽

Natural Combination ◽

Feature Selection Techniques

The method presented in this paper is novel as a natural combination of two mutually dependent steps. Feature selection is a key element (first step) in our classification system, which was employed during the 2010 International RSCTC data mining (bioinformatics) Challenge. The second step may be implemented using any suitable classifier such as linear regression, support vector machine or neural networks. We conducted leave-one-out (LOO) experiments with several feature selection techniques and classifiers. Based on the LOO evaluations, we decided to use feature selection with the separation type Wilcoxon-based criterion for all final submissions. The method presented in this paper was tested successfully during the RSCTC data mining Challenge, where we achieved the top score in the Basic track.

Download Full-text

National and Provincial Profiles of Chemical Releases by Canadian Industrial Sectors, 2002-2010

Inquiry@Queen's Undergraduate Research Conference Proceedings ◽

10.24908/iqurcp.9171 ◽

2018 ◽

Author(s):

Savita Rani

Keyword(s):

Health Hazard ◽

Activity Level ◽

Risk Scores ◽

Regional Health ◽

Industrial Facilities ◽

Industrial Sectors ◽

National Trends ◽

Chemical Releases ◽

National Profile

The National Pollutant Release Inventory (NPRI) is a public-domain record of chemicals released into air, water and land by Canadian facilities from various industrial sectors. The aim of this study was to use historical NPRI data (2002-10) to build national and provincial profiles showing amount, identity and health-hazard classification of chemicals released by facilities in different sectors. Nationally, it was found that 97% of total chemical releases were released into air, and that the top 3 chemical-emitting sectors – Manufacturing (MAN), Mining (MIN) and Utilities (U) – accounted for 98% of these air emissions. Statistical analysis was used to compare provincial chemical releases in the above 3 sectors. Testing showed that significant variation exists in the activity level of the national top 3 sectors within each province. This is reflected in the finding that provincial top 3 sectors do not necessarily match the national profile. Next, health-hazard classifications were determined for the 10 highest-emitted chemicals in the provincial and national top 3 sectors. In the national profile, MAN was classified as carcinogenic, neurotoxic, respiratory-toxic; MIN as reproductive-toxic, respiratory-toxic; U as respiratory-toxic. Sector-hazard relationships in the provinces differed from the national trends and from each other. Ultimately, associating sectors with particular hazards may help link the nature of regional health outcomes to the hazard type of local industrial facilities. A next step would be to account for differing toxicity levels among chemicals of the same hazard type by normalizing the data with risk scores that take into account a chemical’s specific toxicity.

Download Full-text

Classification of Tooth Shapes for Human Identification Purposes–An Experimental Comparison of Selected Simple Shape Descriptors

Lecture Notes in Computer Science - Image Analysis and Recognition ◽

10.1007/978-3-319-20801-5_18 ◽

2015 ◽

pp. 169-177 ◽

Cited By ~ 1

Author(s):

Katarzyna Gościewska ◽

Dariusz Frejlichowski

Keyword(s):

Human Identification ◽

Experimental Comparison ◽

Shape Descriptors ◽

Simple Shape

Download Full-text

Perceived and actual arm performance in multiple sclerosis: relationship with clinical tests according to hand dominance

Multiple Sclerosis Journal ◽

10.1177/1352458513475832 ◽

2013 ◽

Vol 19 (10) ◽

pp. 1341-1348 ◽

Cited By ~ 29

Author(s):

Ilse Lamers ◽

Lore Kerkhofs ◽

Joke Raats ◽

Daphne Kos ◽

Bart Van Wijmeersch ◽

...

Keyword(s):

Multiple Sclerosis ◽

Handgrip Strength ◽

Real Life ◽

Activity Level ◽

International Classification Of Functioning ◽

Fine Motor ◽

Control Group ◽

Hand Dominance ◽

Clinical Tests

Background: The real-life relevance of frequently applied clinical arm tests is not well known in multiple sclerosis (MS). Objective: This study aimed to determine the relation between real-life arm performance and clinical tests in MS. Methods: Thirty wheelchair-bound MS patients and 30 healthy controls were included. Actual and perceived real-life arm performance was measured by using accelerometry and a self-reported measure (Motor Activity Log). Clinical tests on ‘body functions & structures’ (JAMAR handgrip strength, Motricity Index (MI), Fugl Meyer (FM)) and ‘activity’ level (Nine Hole Peg Test (NHPT), Action Research Arm test) of the International Classification of Functioning were conducted. Statistical analyses were performed separately for current dominant and non-dominant arm. Results: For all outcome measures, MS patients scored with both arms significantly lower than the control group. Higher correlations between actual arm performance and clinical tests were found for the non-dominant arm (0.63–0.80). The FM (55%) was a good predictor of actual arm performance, while the MI (46%) and NHPT (55%) were good predictors of perceived arm performance. Conclusions: Real-life arm performance is decreased in wheelchair-bound MS patients and can be best predicted by measures on ‘body functions & structures’ level and fine motor control. Hand dominance influenced the magnitude of relationships.

Download Full-text

Classification of Cancer Recurrence with Alpha-Beta BAM

Mathematical Problems in Engineering ◽

10.1155/2009/680212 ◽

2009 ◽

Vol 2009 ◽

pp. 1-14 ◽

Cited By ~ 2

Author(s):

María Elena Acevedo ◽

Marco Antonio Acevedo ◽

Federico Felipe

Keyword(s):

Breast Cancer ◽

Stable State ◽

Cancer Recurrence ◽

Breast Cancer Surgery ◽

Associative Memories ◽

Training Set ◽

Perfect Recall ◽

Alpha Beta ◽

Leave One Out

Bidirectional Associative Memories (BAMs) based on first model proposed by Kosko do not have perfect recall of training set, and their algorithm must iterate until it reaches a stable state. In this work, we use the model of Alpha-Beta BAM to classify automatically cancer recurrence in female patients with a previous breast cancer surgery. Alpha-Beta BAM presents perfect recall of all the training patterns and it has a one-shot algorithm; these advantages make to Alpha-Beta BAM a suitable tool for classification. We use data from Haberman database, and leave-one-out algorithm was applied to analyze the performance of our model as classifier. We obtain a percentage of classification of 99.98%.

Download Full-text

Possible factors for ankle fractures

Vojnosanitetski pregled ◽

10.2298/vsp1003225t ◽

2010 ◽

Vol 67 (3) ◽

pp. 225-228 ◽

Cited By ~ 2

Author(s):

Dejan Tabakovic ◽

Radovan Manojlovic ◽

Marko Kadija ◽

Mihailo Ille ◽

Goran Turkovic ◽

...

Keyword(s):

Body Weight ◽

Significant Risk ◽

Ankle Fractures ◽

Type B ◽

Physical Exercises ◽

Appropriate Treatment ◽

Fracture Body ◽

Type C ◽

Fracture Types

Background/Aim. Classification of ankle fractures is commonly used for selecting an appropriate treatment and prognosing an outcome of definite management. One of the most used classifications is the Danis-Weber classification. To the best of our knowledge, in the available literature, there are no parameters affecting specific types of ankle fractures according to the Danis-Weber classification. The aim of this study was to analyze the correlation of the following parameters: age, body weight, body mass index (BMI), height, osteoporosis, osteopenia and physical exercises with specific types of ankle fractures using the Danis-Weber classification. Methods. A total of 85 patients grouped by the Danis-Weber classification fracture types were analyzed and the significance of certain parameters for specific types of ankle fractures was established. Results. The proportion of females was significantly higher (p < 0.001) with a significantly higher age (59.9 years, SD ? 14.2) in relation to males (45.1 years, SD ? 12.8) (p < 0.0001). Type A fracture was most frequent in the younger patients (34.2 years, SD ? 8.6), and those with increased physical exercises (p = 0.020). In type B fracture, the risk factor was osteoporosis (p = 0.0180), while in type C fracture, body weight (p = 0.017) and osteoporosis (p = 0.004) were significant parameters. Conclusion. Statistical analysis using the Danis-Weber classification reveals that there are certain parameters suggesting significant risk factors for specific types of ankle fractures.

Download Full-text