Terrorism was still becoming gripping rumours.
After the explosion incident happening or other, police could be it was confirmed busy appointing the list of the suspect and sometimes to begin to spread the photograph of the face, looked for the person who was expected as the perpetrators.
In identified the face of the perpetrators or the fugitive, one of his methods was by comparing the face of the suspect with the photograph archives.
The problem became complicated if the photograph archives were very big and the process of the search was carried out manually.
This method needed time for quite a long time.
The equipment was developed by paying attention to the special characteristics fitur the face of Indonesians.
According to the Blessing, the system was developed by using the method eigenface and the distance euclidean.
The method eigenface was used to carry out the extraction of the characteristics of the important face.
Eigenface based to the Principal component analysis (PCA), the approach that was counted most may always be successful extracted face information.
This method of using the projection of space of the face image with the high dimension to characteristics space with the dimension was lower.
The distance euclidean was used to measure the value of the resemblance between two face images.
It was increasingly small that the clearance of two face images, increasingly high his value of the resemblance.
The picture of the face was in the database extracted then was kept. "When the user put the picture" of the "face that wanted to be determined, the system will look for the similarity between input the data and the picture" of the "available face in the database," said the Blessing that got the degree of the master and the doctor from Tokyo Institute of Technology in the field computional intelegent that.
This system could find the face image that was relevant towards the input image with the level of the accuracy in general 87 percent towards the database of the face image that was used.
The process of the extraction of the characteristics of the face was considered to be simple, fast, and efficient.
The blessing said, the software could be connected with the camera that afterwards automatically was based on the picture that was produced, the system will search whether the face was in the black list or the fugitive.
The equipment will be very beneficial for the world of police.
The identification implement of this face strategic was placed in airports, especially the flight overseas, that usually through the checking table of the document.
The same implement will be very beneficial if being used in police''s post to know the faces of the perpetrators of the crime.
In the Indonesian University personally, the Blessing was discussing with the side of acceptance of the new student of the selection committee how to use the implement.
The use of the implement especially to avoid the practice of the jockey.
Had "had the case in the form" of the "registration" of the "photograph" of "his face the woman, but during acceptance that came evidently the man."
With the implement, could be compared by the similarity of the face, he said.
In the laboratory stage and the prototype, the software could be used.
"For the database, we took the example" of the "face of 500 students the Indonesian University and other pictures from the internet."
Now remained bersinergi with the side that needed him, he said. Without help of the illustrator In his development, the laboratory that was led by the Blessing also developed Cerdas Software for the Identification of the Face of the Fugitive with fuzzy similarity measure.
The program was part of the Research of Unggulan Universitas Indonesia, Program Unggulan and still in the execution.
The identification of the face in police usually with the production of theketch of the face by the illustrator is based on the witness''s information.
The sketch then was compared with the photograph database of the face in police.
The two stages had the big problem.
At the first stage, the sketch accuracy of the face really depended on the interpretation capacity of the illustrator and the verbal capacity the witness in revealing the characteristics of the face.
In the second stage, the official of the seeker of the photograph could in the database experience saturation because of must compare the sketch of the face with all the available photographs.
Software that was developed by the Blessing used antarmuka user friendly that could be used to memvisualisasi the sketch of the face of the fugitive without help of the illustrator.
The user and the witness could at once make the sketch of the face without help of the illustrator.
They were enough to put the verbal characteristics of the face that will be identified by means of choosing property from each characteristic of the face that will be used as the seeker''s component.
To more guided the user in choosing the characteristics, for each choice of property will be given an illustration took the form of the sketch from the characteristics of the face that was chosen.