Draft written Author : Ruhul Amin Parvez
ID: 173-15-10419
Embedded Files
Paper One:
Title: Digital Image Processing Techniques for Object Detection From Complex Background Image.
Authors: R. Hussin*, M. Rizon Juhari, Ng Wei Kang, R.C.Ismail, A.Kamarudin
Time: 2012
Objective: This Paper goal was to detect and allocate the object. They are using a few methods such as color processing and shape detection. The MATLAB program should automatically detect and count the total number of objects based on a scenario. In this paper they choose mango as an object.
Method: In the section of methodology first, they set a default RGB value of the object(Mango) and then set a RGB image adjustment for comparable images. Detect the color of each pixel and determine whether mango color or background which is unrelated to the object color is. Delete the unrelated region by replacing the color with all black color. If the detected color is not related then change the current pixel RGB value to 0 which is in black color.
After the color part is done then the only left image is the clear mango or leaves. Because mango and leaves have the same colors. In the shape detection part they use to detect shape after they complete the color part. For shape detection the image must be changed to grayscale to perform a medium filter to eliminate the small pixel and smoothing the image to make the object's edge more clear and clean.
It will change the image as a binary image which is less than 200 pixels in a group of objects.After that they applied the CHT(Circular Hough Transform) on the selected image to find the circular patterns within the image.
Result: The readable images are very high percentages and it may achieve 100% of detection of targeted objects if combined with other features. Different lighting conditions may affect the colors of the targeted object change not linearly. This project may need to detect the inconstant lighting conditions and complex background, the method of object detection may require to focus on the shape detection by using CHT.
In this paper there are two targeted images. Image number one accurate object detection percentage rate is 60% and image number two accurate object detection percentage rate is 55%
Drawback: In the processing of detection, back light conditions affect the image the most where the RGB color reflected will be different according to the light intensity and its effect on the color processing result as well. Besides that, grayscale images will contain lots of noise because the grayscale filtering cannot eliminate the low intensity pixel which is declared as noise.
Lastly, the CHT may not exactly detect the circular object as sometimes it is connected with other objects together and gives an inaccurate result.
Paper two:
Title: A Model for Intelligent Tourism Guide System
Authors: H.H. Owaied, H.A. Farhan, N. Al-Hawamdeh and N. Al-Okialy
Time: 2011
Objective: Tourist Guide jobs are not easy in modern days. Tourists are often experienced travelers and becoming more and more demanding. As new origins and destinations, areas are opening up frequently. So, the task of a tourist guide is important day by day. For these essentialities an intelligent guiding tourist system was created, which focused on the quality of the visitor’s tourism experience. In this paper intelligent tourist guide system means an intelligent system which is able to provide a greate tourism experience to the travellers. When tourists enable the system it will scan the scenario and it provides information about the place like the real tourist guide.
Method: There are three main types of tour-guide: Local, Rational, National Guide. Local guides work at a particular site, Rational guides work in a rational way, and National guides work nationally.
Those tour-guide follows some commonly used characteristics:
Very well knows the historical region
Some time knows many language either spoken or body language
Knows many stories related with the region
Helpful and good common sense
To implement the intelligent tourism guide, it is necessary to gather knowledge like such peoples.
For this reasons, we have to develop a machine that behaves like human being, smart, problem solver and complex situation handler.To achieve those thinks we have follow some steps like,
User Interface: user interface basically simulates the communication facilities available to be used for interaction with the proposed intelligent tourist guide system.
Knowledge Base: It basically represents the repository of knowledge for specific and narrow domains. Many forms are usually used for knowledge based presentation but those are limited such as rule based, semantic nets, frame, logic forms and case base. Human experts have sense, deductive and analogical facilities.
Facilities of GIS and XRM application: GIS is used to display and analyse spatial data which are connected with databases. Connection between spatial data and database is the driving force behind the work of a GIS. Maps are connected with databases and when data is updated the map also gets updated.
GIS applications in tourism provide the following facilities: Determine important and necessary places, historic and tourist places, best suitable hotel, optimum plan for sightseeing place, shortest distance between the chosen place.
Interface Engine: It was playing the most important role because it made connections with every section like user interface, knowledge base, dynamic database as working memory and functionalities of GIS and XRM application. It is hard to implement general problem solving methods for any field or general search technique also.
Result: Knowbase system, Expert system and the help of the application of GIS and XRM integration will enhance the Tourism Guide system.
Drawback: In this paper there are some techniques that are theoretical and no practical implementations are applied in the paper.
Paper three:
Title: Design and Implementation of an Intelligent Tourist Guide System
Authors: Xinling Wanga, * and Chengzhang Hanb
Time: 2016
Objective: The first intelligent tourist guide arose in the middle of the 1990s. At first it had a simple feature that showed the black and white maps only with the basic positioning services but it was used for experimental research mostly. Nokia Research center developed Tellmoris mobile guide which provides two dimensional maps and three dimensional graphics. Lastly, artificial intelligence agent systems provide better contextual information about user needs and locations.
Based on user-centered design principles this paper(iGuider) points to humanization and intelligence based on ARM11, Embedded system, integrated GIS(Geographic Information System), RF modules and multichannel interaction and other modules.
This tour guide has the features of automatic recognition, automatic playback, multilingual, free choice, high quality, stable and practical.
Method: In this paper, they used many modules like: Positioning module, Control module, GIS module, Display module.
The Positioning module and the hand-held terminal both have a radio frequency module, which is responsible for transmitting and receiving ID respectively.For Implement positioning module there are few hardware requirements needed to fill up.
Control modules also have some hardware requirements if those requirements are fulfilled then it will help to produce needed results.
GIS module is an important module for users because with this user’s can control information, query for function, facilities searching also take the shortest path function and other advanced services.
Display module helps with two parts consisting of visual display and auditory display. The visual display provide good quality video information and auditory display provide the commentary, path tips, public facilities reminders, choose some functionalities options user’s can use those option according their personal preference.
Result: Through the test of tourist attractions results show that this system iGuider is reliable, easy to use, it has good user interface and rapid response capabilities. So, it will bring good economic and social benefits.
Paper four:
Title: Computer Vision and Image Processing: A Paper Review
Authors: Victor Wiley1,*, Thomas Lucas2
Time: 2018
Computer vision has been used for many perspectives like digital image processing, pattern recognition, machine learning and computer graphics. Most of the tasks in computer vision are related to the process of obtaining information on events or descriptions, from input scenes (digital images) and feature extraction. The methods used to solve problems in computer vision depend on the application domain and the nature of the data being analyzed.
Computer vision is the combination of image processing and pattern recognition. Image understanding is the output of the process of computer vision.
The primary purpose of Computer Vision is to create models and data extracts and information from images, while Image Processing is about implementing computational transformations for images, such as sharpening, contrast, among others.
One of the significant challenges in their technique is the sensitivity of the parameters, the strength of the algorithm, and the accuracy of the results.
Computer vision works by using an algorithm and optical sensors to stimulate human visualization to automatically extract valuable information from an object.Compared to conventional methods that take a long time and require sophisticated laboratory analysis, computer vision has been expanded into a branch of artificial intelligence and simulated human visualization.
Image processing has to go through the image analysis process like: image image formation, image processing, image segmentation, image measurement, image interpretation.
There are two approaches to the segmentation and retrieval of image data. Segmentation is basically to divide an image into areas that are not overlapping through specific algorithms to estimate an area of the image. The city is a collection of pixels that have the same unique characteristics as color, gray level, texture, and others.
Fundamentally, segmentation has four main stages: Input image, Segmented map before integration, Edge map before integration, Segmented map and edge map after combination, Pixel clustering.
Segmentation has a primary goal to create a resemblance map which is derived from a prominent object detection model or hierarchical segmentation of the input image. The plan is an aggregation model that tries to form a more accurate salience map
Paper five:
Title: Video Image Processing for Moving Object Detection and Segmentation using Background Subtraction
Authors: Anaswara S Mohan, Resmi R
Time: December 2014
Objective: Detection of video streams and images are used in many computer vision applications. Automatic motion segmentation algorithm works with real images but there are several issues that need to be solved like: noise, missing data and lack of priori knowledge, noise is the difficult one.
Method:
Page updated
Report abuse