Face detection of thermal images based on color image processing
This thesis examines and implements the methods for face detection of thermal images based on color image processing. The overall work describes on how thermal images can enhance the performance of thermal scanner used in the International Airport. In order to examine all aspects of the face detecti...
| Main Author: | |
|---|---|
| Format: | Undergraduates Project Papers |
| Language: | English |
| Published: |
2012
|
| Subjects: | |
| Online Access: | http://umpir.ump.edu.my/id/eprint/7673/ http://umpir.ump.edu.my/id/eprint/7673/ http://umpir.ump.edu.my/id/eprint/7673/1/NUR_FARAHANA_BT_MAT_KHAIRI.PDF |
| Summary: | This thesis examines and implements the methods for face detection of thermal images based on color image processing. The overall work describes on how thermal images can enhance the performance of thermal scanner used in the International Airport. In order to examine all aspects of the face detection with a suspicion of fever, the task has been divided into two consecutive steps: Firstly, Color detection for identifying hot temperature regions on skins and secondly, face detection with no identification. Using a thermal image model for the color region, the possible regions in which faces can be narrow down and could be found rationally. Furthermore, trying to do identification on a region that does not include a face is inefficient. Therefore, using a faster and less complex algorithm to do the general face detection is a faster way to identify faces. In this project, color threshold and mapping method are used in order to detect faces. . In particular to come up with methods that will help increase the chances of correct matches, I propose to apply method focusing on color detection that will detect the higher temperature on human's face using image processing. It focuses on thermal image based method which implemented with MATLAB software. Consequently, in order to successfully be able to identify individuals correctly, advance image processing techniques will be used to enhance the detection rate and efficiency by developing graphical user interface (GUI) in MATLAB to achieve better recognition rate. Finally, all the different steps are merged into an integrated system for full face
detection. The system is evaluated based on accuracy and performance. |
|---|