Firstly, the team would like to thank the external supervisors Dry Jonathan Gogh and Mr. Roth Dubbed from ERR for assisting the team to accomplish the project. The team sincerely appreciates all the guidance throughout the duration of the semester especially from the external supervisors who took time off their lunch hour to meet us for further project discussions or to answer to our queries. Our external supervisors have never once declined our request for a discussion despite their hectic schedules. They gave valuable input and helped the team beyond the boundaries and hence, making this Final Year Project successful.
Secondly, the team would like to thank our internal supervisor Dry. Lu Liming for her extensive care and help and for taking up the crucial role as our internal supervisor; meeting us every week without fail as well as closely monitoring our developments, keeping track of what has been completed and the necessary deliverables during the project and followed by her suggestions and providing, valuable suggestions notably during the many occasions when we had ran out of ideas to enable us to progress forward. The team greatly benefited from her insights on certain areas of security, allowing the team to hint further.
The advice also redefined the team’s boundaries on areas of security. Dry Lu also assisted the team with providing the laboratory facilities and a cabinet for the team’s storage purposes. All in all the team is truly thankful for this immensely educational journey and getting to work with the suppressors. Moving on, the team would like to take this opportunity to extend our warmest thanks to our friends for coming down on several occasions during the course of this project, walking past the Kinetic repetitively during our profiling phase and enabling us to test our system despite having a long day.
This goes out to all the wonderful souls who came over to help us from their workplace and after school. The team understands that it can be quite a time consuming and slightly boring process to be profiled and cannot express our appreciation and gratitude anymore. We would like to thank the twenty six people here for their support. The following mentioned are Lecturers from Diploma in Inform Security (DISH): Dry Lu Liming, Mr. Calvin Sick , Mr. Ho Cheek Eng.
Subsequently, the following are students from DISH, Hafiz Muar, Guan Chi Min, Gusting Nanjing Ting, Ouzo Ye, Darryl Humid, Mikhail, KHz Sock Teen, Des Low, Hew Jinn Ye, Nathaniel Chain, Coho Shih Ha, Young Wee Then. The following mentioned are from Diploma in Business IT (DEBIT): Lee Wee Yang, Fan SAA Chining, Winnie Chow, Tan Huh Our, Calais Than, Trivia Limit, Gerald Tan, Axing Wee , Darrel Lo. Lastly from Diploma in Business Administration (DAB): Limit Lie Shank and Jihad. These folks assisted in the profiling phase.
Last but not least, the team would also like to take this opportunity to make a special mention to two lecturers, Mr. Ho Cheek Eng and Mr. Calvin Sick. Despite Lecturer Mr. Who’s hectic schedule, he took the time off to do profiling or us and shared an article regarding the Kinesics ability to detect a person’s blood flow to identify a subject’s current emotion. After mentioning about the article he had read, he expressly went to his office to search and print out the article for the team for our reference. The actions of Mr..
Ho deeply warmed us and encouraged the team to work harder for the project. Lecturer Mr. Sick was another figure who occasionally dropped by on us offering his thoughts and opinions; he was helpful towards the team, during the holidays when the team required a room to for research and testing repose and was unable to find one, without hesitation Mr.. Sick immediately helped the team book a venue. Finally the team would like to thank everyone for their help directly or indirectly towards completing our Anal Year Project.
This project would not have made it this far without all of their guidance, suggestions and help. Abstract This report presents a research on Gait based Authentication using a Kinetic to identify a subject without being intrusive. Gait is an important part of every humans life. Every individual has their own unique gait. An algorithm is set in lace to calculate length of joints together with an angle to form an identification method. It aims to discover if an algorithm can be calculated based on the diverse findings.
A Gait is defined as a person’s walking behavior. This solution is designed to work as a background process. Research has shown that it is indeed possible to create such a program and at a low operational cost too with the aid of Microsoft Kinetic camera. The Kinetic being equipped with Depth cameras and IR cameras will enhance the results of the data enabling the program to provide an accurate and detailed ATA set. In an attempt to evaluate if the program is suitable to be used as a security implementation, various testing and experimentations would be carried out.
This project is also targeted at economical organizations, where a Kinetic may prove useful as a cost effective solution. But most importantly, this program is targeted and proposed for its ability to be non-intrusive. Through experimentations proper documentations are anticipated to be achieved. Profiling is a critical phase in this project. It helps to gather distinctive data to be used for research. The team is hopeful for the profiling scope to be expanded to more subjects.
It has also been demonstrated that certain materials may not be suitable for this program. Recognition accuracy has achieved an estimate of 86. 66% amongst the 15 subjects. Table of Contents Abbreviations & Symbols –6 Introduction Preliminary Findings – 7-8 Purpose and Scope of Gait Authentication using Kinetic Kinetic for Windows Key Benefits of Microsoft Kinetic 9-10 Microsoft Kinesics Common Uses ? 11 Background Related Work AD Models 12-14 Biometrics 15-18 Initial Group Proposal – – 19-22 Main Text
How did the final implementation come about? 23-24 How the Kinetic Gait authenticator program works? 25-28 Additional Information 29-30 Explanation of program calculation and algorithm 31-33 Profiling Phase 34-35 Percentage of Accuracy 35 Experimentation 36-39 Discussion Software Simpleton Skeleton Tracking Openly /C 41 Opening Framework – -41 Sec ritzy Aspect Team’s view on Kinetic -42 How the program can be improved 43 Program Limitations 44 Program Performance What makes it unique? 45 Current Security Risks – 46-47 Difficulties Faced – 48-49 Conclusion 50 Reference 51-54 Appendix Gaunt Chart 55 user Guide – 56-64 Program Listing 65-68 ADMIT Digital Media and Inform Technology DISH Diploma Inform security technology M Meters CM Centimeter SAVE Support vector machine SMS Sequential Minimal Optimization M LAP Multilayer perception IR Infrared SD AD DNA Software Development Kit High definition 3 Dimensional Deoxyribonucleic acid D TWO DIP RAG Etc.
Dynamic Time Warping Dots per inch Red Green Blue Etcetera COSMOS Complementary metal-oxide?semiconductor INTRODUCTION Preliminary findings The ability to produce an actually functioning automated detection system to identify users based on their unique gaits will be both fascinating and incredible new security implementation the world can have. Gait patterns are always unique due to the fact that different subjects always have a different walking behavior or pattern.
In the 21st century where Information Technology is being used comprehensively, authentications are being enforced to protect the integrity of data and confidentiality. Currently physiological biometric security systems which make use of fingerprints and IRIS to uniquely identify a person has raised concerns over an individual’s hysterical privacy as well as its claim to be one of the most secure methods of authentication has now been put forth in recent years. Researchers have proved that fingerprinting and Iris scanners can be easily manipulated by nothing more than simple tools such as cameras and printers.
It is notable that both of the above mentioned are extremely intrusive and not everyone would support and co-operate with the organizations to make them successful. It is essential for a Biometric system which would eliminate or reduce the risk of privacy concerns as well as being complex enough that it does not become o susceptible. The team could explore the enriched possibilities and options through Biometric system in behavioral characteristics. Research from the past have proven and shown that Gaits are hard to hide and imitate.
One of the greatest advantages enables Gait recognition to not require the subjects attention during the implementation phase as the system would run behind the scenes and detect any possible unauthorized users, alerting the management for further action. These are potential advantages that should be heavily considered. Gait is defined as a subject’s unique walking behavior. During the gait processing phase, the team would be recording a short video and from it, marking out several joints/points which are useful. It will total up to about an estimate of 100 frames for three walk cycles.
A walk cycle is defined by left foot up and right foot down. Every single frame will be put into consideration as the distinct gaits will only surface over time. Every joint will provide the team with EX. coordinates which will be used to calculate any specifics needed. Examples of what can be derived are lengths of certain joints and as well as specific angles between elbows. Since humans are able to identify the people around them just via their unique walk patterns, the team will be attempting to explore more on this subject and hopes to be able to use a Kinetic to simulate the detection of a human.
In this report, the team will be making use of Microsoft Kinetic sensor to research and explore more possibilities. The Microsoft Kinetic has already been programmed to be able to detect and calculate EX.. The proposed areas include just the left portion of the human body. Data will be extracted. In this program, the team will be using Processing, a language that is similar to Java. Data for comparisons will be stored in Myself database. The Kinetic will pass on these data to be processed automatically Proposal of calculating largest angle through frames.
Methodology of using certain variables to ensure authentication is accurate. This report will describe how Microsoft’s Kinetic can be implemented to perform authentication. It will also explain the causes and reasons why Kinetic can prove to be effective for identifying a subject. It will cover the codes, profiling and errors faced during the whole process Of implementation. It will include experiments done by the team as well as how the team derived at the anal solution. The scope covered for this Final Year Project will involve learning how to identify a user and profiling these users’ data into Myself.
The process should be done automated without the need of a personal watching the process manually. Enabling this research successful will require the team to make use of a Microsoft Kinetic. The team aims to research more on Gait Authentication using a Kinetic. Limitations to the program will also be pointed out In the recent years, security has raised concerns on how secure are existing implementations in the market. The team will propose a whole new approach y using length of the joints of a subject calculated with a largest angle between elbows to do identification.
This program should be able to run automated. Besides, the program should be able to detect when a person starts walking. The program is expected to extract necessary data and discard the rest to be stored on the database. The Microsoft Kinetic is able to provide precise video recording with additional features such as depth sensing, infrared capabilities and also having a RAG camera which supports full colors up to 1 IPPP HAD video. The Microsoft Kinetic can be used to develop interactive applications. The Kinetic used by the team will be using Software Development Kit 1. 7 (SD 1 . 7).
Microsoft Kinetic are an important part of security researches. There are quite a few of researches out there that use Kinetic to make security programs. The method used to make these programs vary from colors to AD sketching. One of the main reason there are researchers out there that wants to use a Kinetic instead of high end equipment is due to the Kinetic being cost effective. Key benefits of Microsoft Kinetic: I. Improved body tracking This will allow the depth camera to track body movements to be detected. The sensors can track up to a total of six subject’s skeletons at each time.