sensor fingerprint
In North America, a man named E. Henry in the year 1901 has been successful prior to use
fingerprints to identify the dismissal of workers to cope with the double remuneration. Henry
system derived from the pattern of centralized pattern ridge
fingers, toes, especially the index
finger. Classical method of ink and roll
fingerprint cards printed on a ridge which produces a unique pattern for each individual
digit.
This has proved reliable that no two individuals have a pattern similar ridge, ridge patterns are not able to receive the inheritance pattern of the embryo formed ridge, ridge pattern has never changed in life, and only after death can be changed as a result of decomposition. In life, ridge pattern changed only by chance a result, injuries, fire, disease or other causes that are not fair. Requires fingerprint identification of distinctions about the form of papillary circumference was uninterrupted ridge followed by a mapping of anatomic disorders or signs of the same ridge.
There were 7 papillary ridge patterns:
- Loop
- Arch
- whorl
- Tented Arch
- Double Loop
- Loop and Central Pocked
- Accidental
Having Ridge assertiveness double distance from the beginning to the end, as the width of ridges to each other Ends Evading two different directions ridge running parallel to each other less than 3mm. Bifurcation two different directions ridge running parallel to each other less than 3mm. Hook tear ridges; one long ridges are not more than 3mm Fork Two ridges connected by ridges non third longer than 3mm Dot The ridges are no longer than the adjacent ridges Eye tearing ridges and combining again in a 3mm tear ridges Island and did not merge again , less than 3mm and not more than 6mm. Ridge area is attached. Enclosed Ridge ridges are not much longer than 6mm between the two ridges that are not Enclosed Loop determine the recurrence patterns between two or more parallel ridges Specialties Rare ridge formed as a question mark and hook cutters.
Part two parallel ridges in different areas to surround the pattern was called type lines. They can not possibly continue and restrictions in connection with several sensors, they also may appear fragmented. The starting point of bifurcation, or other anatomic pattern of deviations of two forms of lines, it is called delta. It is generally placed directly in front of the line shape bifurcation. Now further by limiting the delta is indicated, the number of crossing ridge in the pattern of the area gave a ridge count.
Computer tomography can detect relative to the points mentioned above or with a free in the xy space. Anatomic characteristics have an orientation or direction. An analysis of the ridge line direction vector changes can yield an average that reflects this orientation.
fingerprint
The distance between the anatomic pattern of ridge and gave a long line of vectors produced by directing the anatomic characteristics. This is dependent on a sensor that mimics the results can be repeated is not bound by the spreading pressure or melting ridge lines The resultant vector orientation and can be coated to provide a template xy.
Produce anatomic templates of the pattern is not tied to the patterns and flexion / the hunchback can be wrong like the different patterns can have similar anatomic characteristics. No two fingerprints are similar because the number of patterns and anatomic characteristics, anatomic characteristics of their own but is a subset that is too small to be the benchmark.
Large volume of fingerprints collected and stored everyday is an extensive application that includes forensics, access control, licensing and registration director. Introduction of persons automatically based on fingerprints requires the input fingerprint is adjusted with a few fingerprints in a database. To reduce the search time and computational complexity, need to classify these fingerprints in a manner that is consistent and accurate, such that the input fingerprint is required to comply only with a subset of fingerprints in the database.
Fingerprint classification is a technique to assign a
fingerprint into some kind of pre-specified that can not be ignored in the literature that can provide an indexing mechanism.
Fingerprint classification can be viewed as a coarse level
fingerprint that unite it. An input
fingerprint is first adjusted at a level roughly on one type of pre-specified and then, at a level which is better, to be compared on a subset database containing
fingerprints of course.
At this time have developed an algorithm to classify
fingerprints into five classes, namely, whorl, right loop, left loop, arch, and tented arch. The algorithm separates the number of ridges that appear in the four directions (0 degrees, 45 degrees, 90 degrees, and 135 degrees) or by filtering the center of a fingerprint with a bank of Gabor Filters. This information is quantized to produce a Fingercode which is used for
classification. This
classification is based on a two-stage which classifies the use of a K-nearest classifer environment in which the first step and a set of neural networks in the second step. The
classification was tested at 4.000 pictures in the NIST-4 database. For the five-class problem,
classification accuracy can be achieved 90%. For the four-class problem (arch and tented arch combined into one class),
classification accuracy can reach 94.8%. By combining the remaining options, the classification accuracy can be increased to 96% for the five-class categorization and 97.8% for the four-class classification when 30.8% the end image.