基于纹理分析的指纹识别系统设计英文文献

基于纹理的指纹识别系统英文文献 外文文献: AUTOMATED FINGERPRINT IDENTIFICATION SYSTEMS(AFIS) Peter Komarinski This book describes the AFIS process in summary and in detail. The following is a brief explanation of the four components of its name. The automation (A) process has eliminated the need for a print classifier to locate fingerprint cards from a file and compare two physical cards. The searchable database is composed of fingerprint (F) images collected from individuals either by using fingerprint cards or by electronic capture using a device similar to a scanner. The identification (I) aspect occurs when the person is fingerprinted and the resulting images are searched against the database of fingerprint images on a local, state, or national database. It is considered a system(S) because it uses computers and software and can interact with subsystems and other identification systems, including other AFIS systems. AFIS applications exist in almost every instance in which a finger image is rolled onto a fingerprint card. AFIS systems are the primary identification tool for virtually every law enforcement agency in the United States and the rest of the world. An AFIS system can be immense, such as the 46 million records held by the Fabio .It can be small, such as when it contains information about only one city or county. AFIS systems may be linked to other databases, even to other AFIS systems, but there are also some AFIS systems that stand alone and effectively do not communicate with any other agency. As more agencies begin working together, the number of AFIS systems connected together will grow. Stand-alone AFIS systems are more likely to join related systems, creating larger networks of fingerprints to search. The technology and applications of AFIS systems are just beginning to emerge from initial development. The scope of this technology has moved from a select few uses to everyday uses. The core of AFIS technology, the computer and related software, progresses on an almost daily basis. In particular, the software that runs AFIS systems improves constantly as companies develop faster, more accurate programs. New markets have emerged in AFIS-related applications as manufacturers carve out niche products. All of these advances, however, continue to rely on a biometric that has been systematically used for over 100 years: the fingerprint. The use of fingerprints as a biometric used for identification of large population groups can be traced back to the 1890s, when Sir Edward Henry promoted a system of classifying the curving friction ridges and the direction and flow of ridges, patterns, and other image characteristics that allowed trained examiners to translate these images into a set of equations that could be understood by any other examiner trained in the rules of classification. The resulting classifications, in turn, dictated how the records were filed for future retrieval and comparison. A new industry emerged based on the ease with which fingerprints could be captured and a uniform method for measuring these images and storing them for future comparisons. AFIS systems search databases for candidates based on these image characteristics. The characteristics include the points where ridges end, the points where they split, the directions that ridges appear to flow, and even dots. The AFIS system translates what a human sees as a picture, selects key features, searches these features against a database, and produces the best match from that database. Figure1: AFIS pattern types These systems are amazingly fast. It takes only a few minutes to capture the ten finger images at a booking station. Within another few minutes, the booking officer can send the images and arrest information to a state identification bureau. The state can determine the identity and return the identity information and criminal history file (known as a rap sheet) in as little as 30 minutes. If it is the first time the subject has been fingerprinted, the event becomes the first entry in the subject’s computerized criminal history. If the search is for a subject charged with a criminal offense, it includes a check of all 46 million records on the FBI database, yet it normally takes less than 2 hours, the same amount of time required to watch two episodes of JAG or the time it takes to read this book, to get the results. In that short time, the subject’s images can be compared with millions of records at the state and federal level with surprising accuracy and speed. It also takes about 2 hours for a latent print examiner to digitally capture the latent finger image found at a crime scene. By using photographic techniques and software, the latent print image can be made to appear more distinct as the image background is muted. AFIS coders extract the image characteristics from the print, such as location of ridge endings, bifurcations, and direction of ridge flow, and search all or any part of a criminal database. Databases containing millions of image records can be completely searched within minutes. This was not possible just a few years ago. Figure2:Latent Print Process Overview Not all AFIS systems are identical. Some large metropolitan areas have their own independent AFIS system that may or may not directly connect to the state identification bureau. The databases may be mutually exclusive or may overlap. The state AFIS system may come from a different vendor than a metropolitan area’s AFIS, and one vendor’s software may not seamlessly interact with another’s. For example, some systems store images from the two index fingers, some use the two thumbs, and others use a combination. In addition, some AFIS systems provide only identification information and are not connected to a computerized criminal history file. And not all AFIS systems operate on a round-the-clock schedule. Data entered into the database may not be immediately available if the database is updated only once a day. Yet in spite of these differences, the various AFIS systems have a great amount of commonality. They require the same maintenance that other computer systems require, and are subject to the same threats to security and database corruption that other information systems share. Today, more image information, such as palm images and mug shots, are being captured and stored on AFIS systems. A single palm image may have as much ridge detail as that found in all ten fingers. Latent palm prints are estimated to be found at 30% of all crime scenes. Mug shots are used in photo arrays of suspects, and also help visually identify persons who are wanted. These are relatively new capacities made possible by better and less expensive data storage and transmission. In addition, more categories of people, such as health care workers, are being fingerprinted. These new information sources and fingerprintable categories lead to more extensive data-processing requirements, and to the increased responsibility of AFIS managers and technicians, who are handling increasingly larger and more complex systems. While not everyone in the United States is enrolled in a fingerprint-based identification system, images from an inquiry can be compared against perhaps over 50 million records. With the U.S. population at just less than 300 million, that means that one in every six residents of the United States has a record on an AFIS database. That is a lot of records that must be maintained to accurately and reliably produce search results. AFIS systems were developed as a result of the government’s need for prompt accurate identification and industry’s response to that need. The response, however, was not uniform, because standards did not exist in the early years of AFIS. Many large identification bureaus that pioneered the development of AFIS systems found that some of their services were not interchangeable with other AFIS systems, leading to challenges that are still being addressed today. The AFIS process might have never reached its current level of development had not the federal government initiated two important programs that advanced AFIS systems to their current level. First, the adoption of national transmission standards for communication with the FBI provided a“single sheet of music” to sing from. Second, a massive federal funding program for state identification bureaus through the National Criminal History Improvement Program(NCHIP)paid for that“ sheet of music” and the band that plays it. In addition, the introduction and widespread use of computers in the 1980s found a direct application in the field of identification. The infusion of millions of federal dollars, primarily through the NCHIP, combined with a federal presence in the development of standards for transmissions and image capture produced a strong formula for success. The largest AFIS system in the United States is the Integrated Automated Fingerprint Identification System (IAFIS), operated by the Criminal Justice Information Services (CJIS) division of the FBI. The creation of IAFIS became the impetus for new communication and identification strategies. The criminal history database of the FBI found a new home when it moved from the Hedger Hoover building in Washington, DC, to Clarksburg, West Virginia. IAFIS is the national linchpin to which identification bureaus are connected. IAFIS is also the conduit for states to obtain information from other state criminal history and wanted files. The development of AFIS systems has not been restricted to the United States. Several countries in Central America, the Middle East, Asia, and Africa require that all their adult citizens be fingerprinted, and AFIS systems are used to confirm these identifications. In these countries, AFIS may play a role in determining eligibility for government benefits. It can also be used to ensure that persons do not exceed their lawful allocation of goods such as social services benefits and services such as voting. AFIS systems are also used in military applications. Without obvious clues such as a military uniform, it can be increasingly difficult to tell friend from foe, e.g., distinguishing a civilian trying to protect a family in a war zone from a terrorist. With an AFIS system, latent prints found at bombings or other enemy actions can be compared against a database of known individuals. If there is no match, these same latent prints can be retained in the AFIS in anticipation of a match in the future. Fingerprints have no names, no sextant no nationality. Fingerprints do not lie about their past, or appreciably change over time. Fingerprints are relatively easy and inexpensive to capture either with ink and paper or electronically. Combined with sophisticated technology and a skilled staff, AFIS emerges as a practical identification process. Examples given in the following chapters are considered to be representative of AFIS systems. As with automobiles, there are differences between AFIS systems. Some are small compacts, serving only a single community. Some are large trucks that contain all the fingerprint cards in the state. Some are older, less robust systems; as not everyone drives the newest model of automobile, others are state of the art. Just not every ID bureau has the latest and greatest AFIS system. The pre-AFIS systems worked because of the dedication of the staff and the commitment of government to provide criminal history information as quickly and as accurately as possible. In an age before computers, however, the process was very compartmentalized and somewhat tedious. The advances in AFIS technology cannot be fully recognized without some understanding of the tasks that it replaced and why fingerprints are so important for identification purposes.