International Journal of Advanced Sciences and Business (IJASB)

A survey of Recent Advances in Feature Descriptors used for Visual Object Tracking

Khadija LAAROUSSI — 2021-06-10

Visual object tracking is an important and open research topic in computer vision with a wide field of applications. However, selecting the right relevant features to represent the object remains a challenging problem in the tracking community. This paper aims to present a survey of recent progress and advances in feature descriptors which are used to represent the appearance of tracked objects, and then classify tracking methods into different categories. Difficulties in object tracking can arise due to abrupt object motion, changing appearance patterns of the object and the scene, nonrigid object structures, complicated occlusions (object-to-object and object-to-scene occlusions), presence of distractors, deformation, motion blur, scale variation and camera motion. In this survey, we categorize the tracking methods on the basis of the object representation and feature descriptors used to represent the object, we provide detailed descriptions of representative recent tracking algorithms in each category, and then we examine their positive and negative aspects for each category. In this paper, the feature descriptors have been classified into three main categories such as handcrafted features, deep features and multiple features. Through this survey, we would like to present the recent progress in feature descriptors and identify future trends and ideas for visual object tracking research.

Shape Descriptors for Content Based Image Retrieval

Hassan SILKAN — 2021-06-10

Content-Based Image Retrieval (CBIR) systems have been developed to support the image retrieval based on image properties, such as color, shape and texture. In this paper, we are concerned with shape-based image retrieval. In this context, we propose a five descriptors invariant to geometrical transformations and robust to noise to describe shapes, it is based on the curvature scale space theory and extreme curvature. Several experiments were conducted on the widely used MPEG-7 database. The performance was measured in terms of recall and precision. The obtained results show the promising performance of our methods.

The Collaborative Filtering Algorithm on the Big Data Analytics Context

salma ABAROU — 2021-06-10

In recent years, our society has undergone a radical change in various sectors (E-Commerce, medicine, road traffic management, etc.) thanks to the spread of Internet using.

This change has put researchers and organizations facing a real challenge to manage, maintain, classify and make the best decision from the gigantic flows of data created daily.

From which was born the notion "Collaborative filtering approach" which aims to facilitate decision making based on the experience feedback of Internet users. This approach is used generally the big Data Analytics Algorithm to predict and classify data.

Barriers to Implementing Lean Management in Smes: A Systemic Literature Review

Bouzid EL AMINE — 2021-06-10

There is a consensus that SMEs are the main vehicle for the economic development of countries. Lean is a well-established method of organizational philosophy that allows businesses to improve their operations and start more efficiently with greater value and less waste. In addition, SMEs have more obstacles to innovation in their resources and capacities than large enterprises. This makes adopting lean much more difficult for them. Indeed, research on these barriers, shows that the majority of research focused on large companies and omitted SMEs. Therefore, this study will complement a literature review on the barriers to adopting Lean management in SMEs. In this context, we used 35 articles whose titles, abstracts and keywords were studied separately. These data are collected from specific databases which included: Scopus, web of science and science direct. Thus, an inclusive discussion of the barriers associated with the implementation of Lean in SMEs is established.