FINGERPRINT BASED LICENSE CHECKING FOR AUTO-MOBILES J.Angeline Rubella, M.Suganya 1, K.Senathipathi, 2 B.Santhosh Kumar, K.R.Gowdham, M.Ranjithkumar 3 1 Assistant Professor, K.P.R Institute of Engineering and Technology,Coimbatore. 2 Assistant Professor, V.S.B College of Engineering Technical Campus,Coimbatore. 3 U.G. scholar, K.P.R Institute of Engineering and Technology, Coimbatore. Abstract: Driving license system is a very difficult task for t h e government to m o n i t o r. In t h i s project, all the citizens images will scan and recorded. Whenever a citizen crosses the traffic rules, the police can scan his image and can collect penalty / fine from the defaulter. Using this method, the police can track the history of the driver. This biometric based driving license monitoring system is very easy and convenient to monitor. According ancient Greek scripts BIOMETRICS means study of life. Biometrics studies commonly include fingerprint, face, iris, voice, signature, and hand geometry recognition and verification. Many other modalities are in various stages of development and assessment. Among these available biometric traits Finger Print proves to be one of the best traits providing good mismatch ratio and also reliable. Registering the attendances of students has become a hectic work as sometimes their attendance may be registered or missed. To overcome this problem i.e. to get the attendances registered perfectly we are taking the help of two different technologies viz. EMBEDDED SYSTEMS and BIOMETRICS. Index T erms : Authentication, Fingerprint, License, Matching, Minutiae, Sensor. When this module is interfaced to the microcontroller, we will be using it in user mode. In this mode we will be verifying the scanned images with the stored images. When coming to our application the images of the citizens will be stored in the module with a unique id. Citizens have to scan their image on demand by police, which is then verified with the image present in fingerprint module and their record will be updated. This scanner is interfaced to 8051 microcontroller through max232 enabling serial communication. By using this controller we will be controlling the scanning process. After the scanning has been completed the result is stored in the microcontroller. By simply pressing a switch we can get the details of the polling. This project uses regulated 5V, 500mA power supply. 7805 three terminal voltage regulator is used for voltage regulation. Bridge type full wave rectifier is used to rectify the ac output of secondary of 230/12V step down transformer. 2. Block Diagram: 1. Introduction Recently while us discussing about Biometrics we are concentrating on Fingerprint scanning. For this we are using FIM 3030N high voltage module as a scanner. This module has in-built ROM, DSP and RAM. In this we can store up to n no of users fingerprints. This module can operate in 2 modes they are Master mode and User mode. We will be using Master mode to register the fingerprints which will be stored in the ROM present on the scanner with a unique id.
3. Sensor: In this we are using U 4000B sensor for getting the Fingerprint image and to store that in the database. It is an excellent fingerprint input device can be widely applied in social security, public security, attendance, fingerprint encryption, embedded, and many other applications. U 4000B miniature fingerprint scanner to automatically read the fingerprint image, and through USB interface to transfer digital fingerprint images to the computer-controlled technology to support the Biokey SDK development tools. Require authentication for laptop computers, desktop computer or other personal computing devices, it is the ideal accessory. 3.1. Sensing: U 4000B Fingerprints can be sensed using numerous technologies. The traditional ink and paper method, still used by many law enforcement agencies, involves applying ink to the finger surface, rolling the finger from one side of the nail to the other on a card, and finally scanning the card to generate a digital image. In the more popular live-scan method, a digital image is directly obtained by placing the finger on the surface of a fingerprint reader as shown in Figure 2. Optical sensors based on the frustrated total internal reflection (FTIR) technique are commonly used to capture live-scan fingerprints in forensic and government applications, while solid-state touch and sweep sensors silicon-based devices that measure the differences in physical properties such as capacitance or conductance of the friction ridges and valleys dominate in commercial applications. Latent fingerprint impressions left at crime scenes require manual lifting techniques like dusting.3 The most significant characteristics of fingerprint readers are their resolution and capture area. The standard fingerprint image resolution in law enforcement applications is 500 pixels per inch (ppi), but some readers now have dual-resolution capability (500 and 1,000 ppi). The sensing surface of readers used by law enforcement tends to be large so that they can capture palm prints and all four fingers simultaneously such sensors are referred to as 10-print scanners. Low-resolution and small-area readers are preferred in commercial applications so that they can be easily embedded in consumer devices. Sweep sensors are popular in mobile phones, PDAs, and laptops because of their small size (for example, 14 mm 5 mm) and low cost (under $5). However, such sensors require users to sweep their finger across the sensing surface; the reader fuses overlapping image slices obtained during sweeping to form a full fingerprint. Fingerprint sensors embedded in mobile phones or PDAs are also used to support navigation and hot-key functions, with each finger assigned to a specific functionality. 3.2. Feature extraction: Features extracted from a fingerprint image are generally categorized into three levels, as shown in Figure 3a. Level 1 features capture macro details such as friction ridge flow, pattern type, and singular points. Level 2 features refer to minutiae such as ridge bifurcations and endings. Level 3 features include all dimensional attributes of the ridge such as ridge path deviation, width, shape,
pores, edge contour, and other details, including incipient ridges, creases, and scars. Level 1 features can be used to categorize fingerprints into major pattern types such as arch, loop, or whorl; level 2 and level 3 features can be used to establish a fingerprint s individuality or uniqueness. Higher-level features can usually be extracted only if the fingerprint image resolution is high. For example, level 3 feature extraction requires images with more than 500-ppi resolution. Figure 3b shows the flow chart of a typical minutiae feature extraction algorithm. First, the algorithm estimates the friction ridge orientation and frequency from the image. 4. How the fingerprint is recognized and stored in the database: During the enrollment phase, the sensor scans the user s fingerprint and converts it into a digital image. The minutiae extractor processes the fingerprint image to identify specific details known as minutia points that are used to distinguish different users. actual number depends on the size of the sensor surface and how the user places his or her finger on the sensor. The system stores the minutiae information location and direction along with the user s demographic information as a template in the enrollment database. During the identification phase, the user touches the same sensor, generating a new fingerprint image called a query print. Minutia points are extracted from the query print, and the matcher module compares the query minutia set with the stored minutia templates in the enrollment database to find the number of common minutia points. Due to variations in finger placement and pressure applied on the sensor, the minutia points extracted from the template and query fingerprints must be aligned, or registered, before matching. After aligning the fingerprints, the matcher determines the number of pairs of matching minutiae two minutia points that have similar location and directions. The system determines the user s identity by comparing the match score to a threshold set by the administrator. 5. How the database accept the Fingerprint image: Minutia points represent locations where friction ridges end abruptly or where a ridge branches into two or more ridges. A typical good-quality fingerprint image contains about 20-70 minutiae points; the (a) (b) (c) (d) At first the Fingerprint image i.e., the Grayscale image (a) is converted into Orientation Field (b) and then into binary
image (c) and at last the minutiae (d) is matched and stored in the database. 6. Matching: A fingerprint matching module computes a match score between two fingerprints, which should be high for fingerprints from the same finger and low for those from different fingers. Fingerprint matching is a difficult pattern-recognition problem due to large intraclass variations (variations in fingerprint images of the same finger) and large interclass similarity (similarity between fingerprint images from different fingers). Intraclass variations are caused by finger pressure and placement rotation, translation, and contact area with respect to the sensor and condition of the finger such as skin dryness and cuts. Meanwhile, interclass similarity can be large because there are only three types of major fingerprint patterns (arch, loop, and whorl). Most fingerprint-matching algorithms adopt one of four approaches: image correlation, phase matching, skeleton matching, and minutiae matching. Minutiaebased representation is commonly used, primarily because Forensic examiners have successfully relied on minutiae to match fingerprints for more than a century, Minutiae-based representation is storage efficient, and expert testimony about suspect identity based on mated minutiae is admissible in courts of law. The current trend in minutiae matching is to use local minutiae structures to quickly find a coarse alignment between two fingerprints and then consolidate the local matching results at a global level. This kind of matching algorithm4 typically consists of four steps, as Figure 4 shows. First, the algorithm computes pair wise similarity between minutiae of two fingerprints by comparing minutiae descriptors that are invariant to rotation and translation. Next, it aligns two fingerprints according to the most similar minutiae pair. The algorithm then establishes minutiae correspondence minutiae that are close enough both in location and direction are deemed to be corresponding (mated) minutiae. Finally, the algorithm computes a similarity score to reflect the degree of match between two fingerprints based on factors such as the number of matching minutiae, the percentage of matching minutiae in the overlapping area of two fingerprints, and the consistency of ridge count between matching minutiae.
6.1. How it matches with the stored image 6.1.1. Sensor s identification process: 6.1.4. Work of matching process: 6.1.2. Recognition Algorithm: 6.1.3. Image Processing: This part consists of six stages. At the image enhancement stage, noise on the input fingerprint image is eliminated and contrast is fortified for the sake of successive stages. At the image analysis stage, area where fingerprint is severely corrupted is cut out to prevent adverse effects on recognition. The binarization stage is designed to binaries a gray-level fingerprint image. The thinning stage thins the binaries image. The ridge reconstruction stage reconstructs the ridges by removing pseudo minutiae. At the last stage, minutiae are extracted from the reconstructed ridge image. Matching stages show big differences according to their types although they are based on the same minutiae. Here, the most well-known matching algorithm will be briefly explained. The matching process consists of four main stages. First of all, the minutiae analysis stage analyzes the geometric characteristics such as distance and angle between standard minutiae and its neighboring minutiae based on the analysis of the image-processed feature data. After the analysis, all the minutiae pairs have some kind of geometric relationship with their neighboring minutiae, and the relationship will be used as basic information for local similarity measurement.
7. How to over come Duplication: 7.1. Altered/fake fingerprints: People may alter their fingerprints in different ways for many reasons. For example, an unauthorized user may use a fake finger that imitates a legitimate user s fingerprint template to access a computer system. Criminals may cover their fingers with fake fingerprints made of substances like glue or they may intentionally mutilate their fingers to avoid being identified by automated systems or even human experts. An essential countermeasure to thwart the use of inanimate or fake fingers is liveness detection checking if the finger is live by measuring and analyzing various vital signs of the finger such as pulse, perspiration, and deformation. While software-based liveness detection solutions that complement existing fingerprint scanners may be more costeffective, they have not yet shown much promise. To deal with mutilated fingers, a mutilation detector should be added, and, once mutilation is detected, effort should be made to identify the subject either by restoring the original fingerprints or using only the unaltered areas of the fingerprint. With the adoption of multiple biometric traits in largescale identification systems such as the FBI s NGI, multi-biometrics will be a powerful tool to handle altered fingerprints. Duplication can be avoided since each and every Fingerprint has it s uniqueness in nature 7.2. Uniqueness of the Fingerprint: The "fingerprint" which is formed on the tip of the finger by the visible pattern the skin takes is absolutely unique to its owner. Every person living on the earth has a different set of fingerprints. All the people who have lived throughout history also had different fingerprints. These prints remain unchanged throughout one's lifetime unless a great injury occurs. That is why the fingerprint is accepted as a very important identity card and used for this purpose around the world. However, two centuries ago, the fingerprint was not so important, because it was only discovered in the late 19th century that all fingerprints are different from one another. In 1880, an English scientist named Henry Faulds stated in an article published in Nature that the fingerprints of people did not change throughout their lives, and that suspects could be convicted by the fingerprints they left on surfaces such as glass.32 In 1884, for the first time a murder was solved by means of identifying fingerprints. Since then, fingerprints have become an important method of identification. Before the 19th century, however, people most probably had never thought that the wavy shapes on their fingertips had any meaning or considered them worthy of note. In the 7 th century, the Qur'an pointed out that the fingertips of human beings bore an important characteristic. 7.3. How to overcome for injured fingerprint: Thus, the accidental injuries are common. The injured fingerprint will get its own true image after it get cured. But for our security purpose we should take 2 images of the fingerprint from 2 different fingers. 7.4. Here are some general tips: Because it is possible to damage one or all the fingers on one hand, users are encouraged to enroll at least one finger from each hand. Careful enrollment is essential. Once a fingerprint is enrolled, its template will be used for future authentication, and careful enrollment helps obtain better (beyond the
threshold quality) templates and hence will help to avoid false matches in the future. Attention should be paid to the sensor surface as the eye of the system. It should not be touched or scratched with dirty fingers or anything hard. If the sensor is dirty, wet, or fails to work, it should be gently cleaned with a dry, soft, lint-free cloth. If someone has difficulty enrolling or authenticating, the hands should be cleaned or a different finger used. If the hands are too dry, lotion can be applied, but not too much because that can have the opposite effect. If appropriate, the user might rub the finger to be matched on his or her forehead to moisten it with the skin s natural oil. 8. Suitability of fingerprint recognition: 8.1. Universality: Only very few people miss all 10 fingers. Most fingerprint recognition software allows enrolling multiple fingers which avoids that an individual is no longer granted access after injury. 8.2. Uniqueness: It is generally accepted that fingerprints are unique to an individual. However, there is a risk that fingerprints of two different individuals match if the fingerprint image is of insufficient quality. Therefore the False Acceptance Rate (FAR) is highly dependent on the quality of the fingerprint reader. 8.3. Permanence: Fingerprints do not change with ageing, but as people age they lose collagen which makes their fingerprint harder to read and this can lead to significantly more false rejects with elderly people. Injuries, such as fire wounds, can damage a fingerprint but if multiple fingers are enrolled the likelihood of an authorized individual being denied access is reduced. 8.4. Collectability: Fingerprints are easy to acquisition, the cheapest fingerprint readers available use a digital camera. Fingerprint readers that are more difficult to fool, such as CMOS readers, are even not overly expensive. In some environments, where for example people are unable to wash their hands, more expensive means might be necessary to acquire a useable fingerprint image. 8.5. Acceptability: Fingerprints are easily accepted as soon as people reflect that they leave their fingerprints everywhere and that no sensitive information, such as medical conditions, can be derived from fingerprints. 8.6. Circumvention: There are a number of concerns when using fingerprint recognition. A finger can be cut off, this is no joke it already happened. Fingerprint sensors with liveliness detection can resolve this issue. Fingerprint dummies are not too difficult to make, the effort is highly dependent of the biometric device to be fooled. Some of the cheapest devices can even be fooled by a fingerprint image that is printed on paper or transparency. Dummies can be created for each type of sensor, however in general the more complicated (and thus expensive) sensors are more difficult to fool. Liveliness detection do make fingerprint readers a lot more difficult to fool. Make a thorough analysis before implementing fingerprint recognition as a means for authentication or identification in a high securityenvironment. Avoid using fingerprint readers without liveliness detection.
8.7. Performance: In terms of speed, accuracy and robustness the devices actually on the market should cover any need, except maybe for big corporations and government applications where matching algorithms might become a bottleneck. 9. Some of the advantages of our proposed Method: 1. Since fingerprints are the composition of protruding sweat glands, everyone has unique fingerprints. 2. They do not change naturally. 3. Its reliability and stability is higher compared to the iris, voice, and face recognition method. 4. Fingerprint recognition equipment is relatively low-priced compared to other biometric system and R&D investments are very robust in this field. 5. Very high accuracy. 6. It is one of the most developed biometrics. 7. Easy to use. 8. Small storage space required for the biometric template, reducing the size of the database memory required 9. It is standardized. 10. No need to carry License 11. It avoids Fraud & Duplication. 12. More economical and effective method. 10. Conclusion: Automated fingerprint identification systems have been successfully deployed around the globe for both law-enforcement and civilian applications, and new fingerprintmatching applications continue to emerge. The fingerprint will continue to be the dominant biometric trait, and many identity management and access control applications will continue to rely on fingerprint recognition because of its proven performance, the existence of large legacy databases, and the availability of compact and cheap fingerprint readers. Further, fingerprint evidence is acceptable in courts of law to convict criminals. In this paper we have proposed method based on Minutiaebased algorithm for efficient and more secured because of these features Universality, Uniqueness, Permanence, Collectability, Acceptability, Circumvention and Performance when compared to the existing system. The security can be further increased using some modern technologies like advanced sensors, e.g. Retina Sensors, but these increases the cost of the project. 11. References: 1. National Science and Technology Council Subcommittee on Biometricsand Identity Management, Biometrics in Government Post-9/11: Advancing Science, Enhancing Operations, Aug 2008. 2. A.K. Jain, P. Flynn, and A.A.Ross, eds., Handbook of Biometrics, Springer, 2007. 3. H.C.Lee and R.E.Gaensslen. eds., advences in FringerprintnTechnology, 2 nd., CRC press 2001. 4. J. Feng, Combining Minutiae Descriptors for Fingerprint Matching, Pattern Recognition, Jan. 2008, pp. 342-352 5. A.A. Ross, K. Nandakumar, and A.K. Jain, Handbook of Multibiometrics, Springer,2006. 6. S. Pankanti, S. Prabhakar, and A.K. Jain, On the Individuality of Fingerprints, IEEE Trans. Pattern Analysis and Machine Intelligence, Aug. 2002, pp.1010-1025.