Identificationof Human based on Dental photograph or Dental radio graph is one of the popularand emerging trends in Biometrics.
The Dental biometric system utilize dentalphotograph and dental radiograph foridentification of person as it provide unique teeth feature such as crown,bridges, fillings, and relative distance between neighboring teeth. Forimplementation of a dental biometric system we have to implement variousTechniques such as Image enhancement, Segmentation, Feature extraction andMatching. In this paper we try to make an effective system that identify personbased on their dental records. Keywords: Antemortem (AM), Dental Biometrics, Dental Anatomy, Post mortem (PM) 1. IntroductionImage processing is amethod to convert image into digital form by performing some operation on it,in order to get some useful information or some enhanced features from it 16.It is a type of Signal processing where input is image and output is image orsome characteristics features associated from it. Now a days image processingis Rapidly Growing technology where lots of research work is done in current trend.
Imageprocessing basically include three basic steps: First is to importing Imageswith the help of Optical scanner or digital photography. Second, is tomanipulate or analyse image by image enhancement or image compression. Third,is output of image . 1Image processing is used for many purpose likevisualization, image sharpening and restoration, Image retrieval, Measurementof pattern , Image recognition. There are two types of image processing methodsDigital image processing and Analog image Processing. We focused on DigitalImage processing Methods and Techniques. Digital image processing refers to use ofcomputer algorithms and techniques to perform image processing on DigitalImages.
An image may be defined as two dimensional function f (x ,y)where x and y are Spatial coordinates and amplitude f at any pair of coordinates (x ,y) is calledintensity of the image. When x , y and the amplitude values of f are allfinite, discrete quantities, we call the image as digital image. 2Biometric derived from the Greek word “bios”for life and “metron” for to measure. A biometric system is used to identifyperson based on their physical characteristics such as finger prints , teethpattern etc.
A biometrics is a convenient and reliable method to identify humanidentity. Biometrics system is divided in to two categories: Behavioural biometricsystem and physiological or physical biometric system. Behavioural biometricsis the field of study related to measure of uniquely identifying and measurablepatterns in human beings such as signature password. physiological or physicalbiometrics is the field of study related to specific characteristics of humanssuch as finger prints, iris patterns, face and palm print patterns. 3 Dental biometrics is also used in forensicidentification of human. Human Identification based on dental information wasfirst observed in Roman Empire when Nero’s mother Agripina ordered to killloilla paulina who was later identified by her dental caries and bad dentalcondition.
The first treatise of human identificationbased on dental records was done in 1857, by Dr Oscar Amoedo Valdes who applieda dental based identification technique to identify disaster victim in Paris.After his work human identification based on teeth is accepted worldwide. 4Duringlast decades human identification based on dental records was extremely used toidentify victims in massive disasters such as 9/11 terrorist attack in New yorkand tsunami in Asia. Dental Biometrics in Forensic identification ofhuman uses AM (anti mortem) and PM (post mortem) records in database formatching Teeth patterns to identify human beings. Biometric features such asfinger print , face is destructed in accident such as plane crash, burningaccidents, floods ,so it is not possible to work with conventional methods soin such case to identifying persons because bones and teeth does not easilydecay and it takes high temperature up to 1100 degree temperature for melting. 5 2. Dental Anatomy Dental Anatomy is a field of Anatomy which deals with study of humantooth structure .
It is also a taxonomical science that deals with naming teeth& structures of teeth fig a. A Tooth (Teeth) is a small calcified whitishstructure found in the jaws (mouth) for breaking food. Tooth formation beginsbefore birth.
Teeth are not made of bones but it is made of various types oftissues, enamel etc. Teeth are hardest substance in human body. Different partsof teeth are fig b: 1) CROWN: It is top part of tooth that wenormally see. Shape of crown determines tooth structure.
2) GUMLINE: In this Gum-line tooth and gums aremet. 3) ROOT: It is a part of tooth which isembedded in bone.4) ENAMEL: It is outermost layer of teeth whichis hardest substance found in human body.5) DENTIN: It is next layer of tooth foundafter enamel.
6) PULP: It is the soft tissue found in the centreof teeth.Usually there are 20 primary (Baby) teeth and28 to 32 permanent teeth, last 4are third molars or wisdom teeth each of whichmay or may not grow in. Fig a shows that there are 16 teeth found in maxilla(upper jaw) and 16 found in mandible (lower jaw). There are four types of teeththey are:1) INCISORS: sharp chisel shaped front teeth (4upper and 4 lower) used for cutting.2) CANINES: they are called cuspids because oftheir shapes like cusps, used for tearing food.3) PREMOLAR: these teeth have two cusps shapefor tearing and grinding.4) MOLARS: These teeth have many cusps on theirbiting surface and use for grinding food Figurea.
Tooth Anatomy Figure b. Structure of tooth 3. Types of DentalRadiograph Dentalradiographs are commonly called as X rays.
A Dental radiographic image isformed by a controlled burst of X ray radiation which penetrates oralstructures at different levels, depending on varying Anatomical densities,before striking the film or sensor. Teeth appear lighter because less radiationpenetrates them to reach the film. Dental caries, infections and other changesin the bone density appear darker because X ray readily penetrate these lessdense structures. Dental restorations (fillings, crowns) may appear lighter ordarker depending on the density of material. There are various types of dentalradio graph few of them are as follows 3: 1)BITEWING: This technique involves biting down a special piece of paper so thatwe can see the crowns of teeth.
See (fig 1) Figure1. Structure of tooth2)OCCLUSAL: This X ray is done when your jaw is closed to see how your upper andbottom teeth line up.See (fig 2) Figure 2. Structure of tooth 3)PANORAMIC: In this type of radiograph the machine rotates around the head.
See (fig 3) Figure 3. Structure of tooth4)PERIPICAL:This technique focus on two complete teeth from root to crown. See (fig 4) Figure4. Structure of tooth 4. Dental Photograph DentalPhotograph See (fig 5) is an image which is taken with the help of camera.
Insome case where dental radiograph is not obtained easily on that case we useDental Photograph. It is easy way to collect Dental Photograph where we cancollect Dental photograph of various persons. For identification of humanDental photograph also play an important role. Use of dental Radiograph inidentification of human takes less cost and it is easily Available. We can useany camera but its resolution is good enough so that it can take good qualityof image.
Use of good quality of dental photograph gives better output and makeour Dental based identification system more effective. In below figure we seethe Dental image of various persons. Figure 5. Structureof tooth 5. Literature Review In this review we present all the previousreview of some selected paper in the form of table. Below table illustrates the literature review ofvarious techniques which is used by different authors in Previous Years. Table 1. review of various techniques Year Reference Enhancement Technique Segmentation Technique Feature Extractor Matching Technique Results 2017 G.
Jaffino, A.Banumati, Ulaganathen Gurunathan, J.Prabin Jose 6 Butterworth High pass filter Bite viewing radiograph using Spline Isolation, Missing tooth detection Moments Feature Extraction K-NN classification & Texture based, Fuzzy classification based matching Efficient Algorithm with 93% accuracy 2015 Soma Datta & Nabendu Chaki 7 De noising winner filter Watershed & snake Algorithm used ROI Extraction such as volume, size, area NA Efficient but not 100% accurate 2015 Dipali Rindhe & Ganesh Sable 9 Assumed good quality images were used in JPEG format Thresholding SIFT ALGORITHM Matching feature based on Euclidean distance It give better matching rate for identification 2014 Vijaya 8 Kumari pushparaj, Ulaganathen Gurunathan, Banumati Arumugan Use good quality of radiograph & photograph Missing tooth Identification & Spline isolation Contour Extraction Binary (SVM) Contour Based Matching This Method Outperforms other methods in terms of Computation time & hit rate 2012 Shubhangi C.Dighe & Revti Shriram3 Histogram equalization Extraction of ROI Thresholding Difference between radiograph based on dental work , mode, median, skewness Control Point Selection of tooth contour Method give direct visualization of matching or not matching & unsatisfactory for two blurred image Table1.review of various techniques cont Year Reference Enhancement Technique Segmentation Technique Feature Extractor Matching Technique Results 2011 C.K Modi 1017 Assumed good quality of images were used Radiograph Segmentation Fourier Descriptors Absolute and Euclidean distance 66.667 % Precision 2009 P.L.
Lin & Y.H.Lai 1117 Hommorphic Filtering ; Adaptive contrast stretching morphological transformations Thresholding ; Horizontal & Vertical integral projection Relative length/width radio of a teeth, Relative crown size Only classification is performed 94.
9% for molars &95.6% for premolars. 2008 S.Kiattisin 1217 Brightness Adjustment ; Binary image conversion Chain code Decoding Special features of teeth Absolute matching between decoding & statistical code Same code match=90%(Same pattern); 50%(different pattern). 2006 O.
Nomir and M. Mottaleb 1317 Binary image masking Radiograph Segmentation: Iterative & Adaptive thresholding ; Horizontal & Vertical integral projection Tooth Contour Hierarchical Chamfer Matching Algorithm Out of 50PM quer images, Rank I-40, Rank II-3, Rank III-2, Rank V-4, Rank VII-1 images. 2004 O.
Nomir and M. Mottaleb 1517 Binary Image Masking Iterative & Adaptive Thresholding : Horizontal & Vertical integral projection Signature vector From tooth contour Absolute distance matching Out of 29 PM query images, Rank I-21, 4 out of %PM images correctly retrieved. 2004 Anil K. Jain 1417 Assumed good quality images were used Horizontal & Vertical integral Projection Contour of Teeth & Shapes of Dental work Computation of image distances matching Out of 11 PM query images,Rank-I-72%, Rank-IV-91%, Rank-VII-100% 6. Proposed Methodology The methods we have proposed herecomprises five main processing steps.All the image will pass these process. First dental radiograph is sent in toimage enhancement technique where we improve the quality of image then image issent to pre-processing stage where we process image then image is segmented insegmentation process then we extract some special feature by use of FeatureExtractor then we match the image to get desired output.
The proposed system isimplemented with five modules:1.Pre-processing 2. ImageEnhancement3. Segmentation4. FeatureExtraction5.
Dental imageMatching Blockdiagram of proposed systemThe block diagram ofdental biometrics is shown in Figure 4.1. Figure 4.1: Block diagram of Dental biometric system Pre-processing Feature Extractor Matching of dental image Output of image Image Enhancement Segmentation Dental radiograph 6.1. Pre-processing It is the first step or fundamentalstep of Digital image processing. Image pre-processing or Image Acquisitionbasically deals with improving the quality of image. If our image quality ispoor then we have to remove noise from that image and make our input image moreaccurate.
For removing Noise various types of Filters is used.Removalof Noise in done in 3 ways: 1. Remove Noise by LinearFiltering. 2. Remove Noise Usingan Averaging Filter and a Median Filter. 3. Remove Noise By Adaptive Filtering.
6.2. Pre processing It is Simplest and mostly applied in variousareas of digital image processing. It is used in those areas where thesubjective quality of images is important for human perception.
The idea behindthis is to highlight certain features of interest in an image.6.3.Segmentation InSegmentation process we partition an image into constituent parts or object. Inthis we segment our dental photograph or radiograph such that each regioncontains only a single toot. There are many different ways to perform image segmentation, including:1.
Thresholding Methods such as Ostu’s method.2. Transform Methods such as watershed segmentation.3. Texture Methods such as Texture Filters.
6.4. Feature Extraction Feature Extraction is a type of Dimensionallyreduction that efficiently represents interesting parts of an image as afeature vector. This is technique is used when image size are large and reducedfeature representation is required to complete tasks such as image matching. Inthis step we extract useful features of teeth so that it helps in matching andclassification.6.5.
Matching After doing all the previous process such aspre-processing, image enhancement, segmentation, feature extraction the laststep is to complete our target is matching of image. For matching we have todevelop an algorithm based on properties applied and image features anddeveloped distance matching. Accurate matching leads good identification whichis our desired goal.7.
Expected Results We expect that ourDental Biometric system is able to Identify Person Based on their Dentalrecords. We use various Techniques and Methodology in our system so that it canidentify person more precisely and accurately. In previous system of DentalBiometrics for Human identification Various Technique is used so our task is touse previous methods and Techniques and adding new Techniques to make oursystem more Effective.8. Conclusion and Future work Dentalbiometrics is not only used for identification of person but it is act as aconfirmatory tool.
It is used in forensic identification of human as well asdisaster victim classification. Among other Biometric Methods for IdentifyingHuman, Dental based Identification (DENTAL BIOMETRIC) gives better results invarious conditions where person body parts were Damage. With the help of aboveresearch in this area, we can make an effective Dental Biometric System withefficient and time saving methods with accurate Results. Our main Challenge andFuture work in Dental Biometric system for human Identification is to deal withpoor quality of images, Teeth overlap, teeth shape change consideration due toageing.