2D Facial Composite through Image Processing Techniques

Abstract

The facial composite is one of the major fields in forensic science that helps criminal investigators to carry out their investigation process. The survey conducted by United States Law Enforcement Agencies confirms that 80% of the law enforcement agencies use computer based automated composites whereas Sri Lanka is still far behind in the process of facial composite with a lot of inefficiencies in the current manual process. Thus, this research involves discovering a new way of carrying out the facial composite process, while eliminating the manual hand drawn procedure that currently prevails in Sri Lanka. Even though there are many global software, the usage of these software does not assist in creating 2D facial composites in Sri Lanka as the facial feature templates do not look alike to the local population. Thus this research introduces the image processing based computer automated software solution with 2D facial feature templates targeting the Sri Lankan population. None of the previous research studies have been used for the approach of constructing composite images combining the two major sectors known as medical field and aesthetic field. This is the first ever approach that creates the 2D facial feature templates by incorporating both medically defined indexes and aesthetic aspects. Furthermore , the research also consolidates a detailed rule set based decision support system which has been implemented for the first time to filter the most possible facial feature templates according to the gender, age, height, weight (BMI- Body Mass Index) and other criteria given by the eyewitness. Hence these 2D facial feature templates have created, analyzing the sample population that ranges between the 20 -25 year age categories. Two sub researches have been carried out to identify the index ranges and the shapes for both male and female. The outcome of these two sub researches has been used in the creation of 2D facial feature templates for Sri Lankan population. According to the Face Pool evaluation technique, an average of 70% success rate has been obtained from the ultimate composite images from the proposed process. Henceforth, dramatic increment of the success rate emphasized the effectiveness of the process introduced in this study. Thus this study will be a solid foundation for the researchers who are expecting to improve the automated facial composite process.