Forensic 3D Laser Scanning of Footwear Impression Evidence - Thesis Proposal Example

3D laser scanning of footwear impression evidence with the use of a FARO "Freestyle" 3D Scanner al affiliation: Introduction It is very critical to capture impression evidence images in crime scene investigations. For instance, shoe impressions, tire tracks, fingerprints, bloodstains and many more are captured (Houck and Siegel, 2010). For a crime investigation to be efficient it depends entirely on collection and analysis of different evidence and information collected. Currently, the evidence is recorded by a two dimensional (2D) photographs which are mainly colored or physically making a cast impression to capture a three dimensional (3D) configuration of the evidence (Houck and Siegel, 2010; McDonald, 1992). However, getting a 3D physical cast of an impression on a crime scene can in the process destroy the evidence. The use of 3D imaging device is proving to be beneficial to crime scene investigators (CSI) as an additional toolkit since it can obtain details of such impression. Footwear impression recovery in crime scene is considered to have a critical role in the investigations to narrow down the number of suspects or refute or corroborate information (Gamage et al., 2013). Three dimension foot ware impressions casting is seen to be a long standing standard where, it results to the generation of three dimensional footprints. This is progressively being replaced by a three dimensional scanning that is a less invasive method. In the study proposed, an alternative method that entails a 3D laser scanning of footwear impression evidence with the use of a FARO "Freestyle" 3D Scanner will be present, and it will be based on the multi-view stereo where, an accurate three dimensional model can be generated. The results on the reconstructed three dimensional models will be evaluated and then compared with other results obtained by three dimensional scanning. Also, the drawbacks and the advantages of this method will be explored over other methods that exist. Freestyle hand held scanner was recently released by FARO, and this is considered as a step up in structured light technology because it includes two cameras infrared in nature which are used in creating a stereo pair of images Objectives Three dimensional foot ware impression casting is progressively being replaced by a three dimensional scanning that is less invasive method. In the study to be undertaken, an alternative method that entails a 3D laser scanning of footwear impression evidence with the use of FARO "Freestyle" 3D Scanner will be present and it will be based on the multi-view stereo. Reconstruction of three dimensional models The results on the reconstructed three dimensional models will be evaluated and then compared with other results obtained by three dimensional scanning. The drawbacks and the advantages of 3D laser scanning of footwear impression evidence with the use of a FARO "Freestyle" 3D Scanner will be explored Research Questions Is the reconstructed three dimensional models evaluated better than other results obtained by three dimensional scanning? What are the drawbacks and the advantages of this method over other methods that exist? Hypothesis Footwear impression recovery in crime scene is considered to have a critical role in the investigations to narrow down the number of suspects or refute or corroborate information. Freestyle hand held scanner was recently released by FARO, and this is considered as a step up in structured light technology because it includes two cameras infrared in nature which are used in creating a stereo pair of images. This offers an improved accuracy and it is seen to be analogous to the way humans see with their both eyes. Contribution The study will involve a 3D laser scanning of footwear impression evidence with the use of a FARO "Freestyle" 3D Scanner. It will be based on the multi-view stereo where, an accurate three dimension model can be generated. The key contributions of the study to be undertaken include obtaining footwear prints, the three dimension models from pictures that are taken at various angles. The result will as well be analyzed based on accuracy and its merits will be presented over other methods. Literature review Casting is seen to have been the major method for three dimension footwear impression evidence for the past several decades. Despite there is photographing of the impressions, it is seen that physical storage space is taken up by complex equipment (DeLaurentis, 2009). It is as well critical to note that creating a detailed as well as complete three dimension model for footprint consumes less than fifteen minutes and this is considered to be faster than the casting method (DeLaurentis, 2009) According to Bodziak (2000) study conducted in Switzerland, it is estimated that shoe prints can be found in about 35% of all crime scenes.Shoe prints at a crime scene can give an impression of whether the person was running, walking, familiar or unfamiliar with the area terrain (Hess and Orthmann, 2010). It gives more information about the person wearing the shoe; they include height, weight, wear patterns of the shoe. With all these information available, they can be compared with those of the suspects. With the help of all the factors put in to consideration and also the location of the shoe print, the investigators can reconstruct the crime. Shoe prints can be put into three categories based on the way they are found at the crime scene. The categories include patent, latent and plastic where, the patent shoe prints are referred to as the prints that are clearly seen at a crime scene (Houck and Siegel, 2010; McDonald, 1992). On the other hand, latent prints are not visible by naked eyes hence the need to express them using different forensic techniques. The plastic or a three dimensional print is one that a shoe sinks in the material(s) being stepped on which leaves marks. Three dimension footwear imprints have depth, length and width. They are mainly found outdoors such as in the snow,sand and soil. There details can be taken, retained and recorded depending on the material composition, texture and condition which they do vary (Houck and Siegel, 2010; McDonald, 1992). 3D scanners use in practice and in forensics is growing hand in hand with the field of Computational Forensics (CF).The emerging of CF as an interdisciplinary domain in research it is understood as hypothesis driven investigation of a given forensic problem with the use of computers. Numerous works are attributed to the field of computational Forensic; they include latent palm print matching, automatic image retrieval of a shoe print, estimation of a person or object height, visualization of crime scene using 3D and many more (Houck and Siegel, 2010; McDonald, 1992).Different forensic techniques are required to reveal, capture and also analyzed each type of evidence collected. Over the last years this techniques have been better and better. This is attributed by reliability of improved modern technology and also largely the use of computational forensics. The use of photos to 3D pipeline technique involves a set of photographs of a footwear impression which are used as an input. The photos are taken at several viewpoints round the evidence, and then a 3D point cloud is generated. The 3D point cloud is used to obtain a 3D surface later on. The use of pattern recognition and additional computation methods helps to decrease prejudice inherent in the traditional criminal forensics. In this case, the ever developing system to collect, organize and also analyze evidence is 3D scanning. The High Definition Surveying (HDS) or 3D scanning has become common in the surveying of terrain, buildings and many other architectural in a more detailed way and faster rate. In forensic there are two methods of 3D scanners applied and include: Crime scene scanners: they capture a larger overview of the crime scene. Close up 3D scanners: They capture individual objects in a high resolution and full color. With close up 3D scanner, shoe prints, bones, footprints, bones, tire tracks and many more can be used in this case (Liscio, 2009). Advantages of 3D scanning over casting The 3D scanners use lasers that scan the evidence objects without affecting it or coming into contact with. On the other hand casting can destroy the evidence in the process as it comes into contact with the evidence. Various scanners capture the clear color surface hence producing accurate replica which is not entail true with casting method.Furthermore, sharing of evidence across board is easier using 3D scanning unlike the casting method. Also the time used to process crime scene evidence using 3D scanner is much less than the use of traditional casting method. The traditional casting method is mainly used to obtain simple evidence with fewer curves unlike 3D scanning which used to scan highly curved organic shaped evidence surfaces. However, both methods produce a high rate of accuracy result. The main disadvantage with 3D scanning is that it not appropriate method to be used in all kind of evidence surface materials. Multiview stereo entail photographs taken of an object at different viewpoints of a camera. They work out correspondences amongst depth and image pair’s approximation of every camera viewpoint. When these estimates are combined, multiview stereo can give final 3D model of a given object. This approach was used to recover footprints of a dinosaur successfully. FARO "Freestyle" 3D Scanner Freestyle hand held scanner was recently released by FARO, and this is considered as a step up in structured light technology because it includes two cameras infrared in nature which are used in creating a stereo pair of images (Eugene, 2015). This offers an improved accuracy and it is seen to be analogous to the way humans see with their both eyes. The point of accuracy stated is approximately 1.5mm, and the scanning range is estimated to be 3m.The Freestyle scanner is seen to have several features form other systems where, this features are integrated in the system. For instance, the sensor is seen to be of high quality as well as robust in nature (Eugene, 2015). The optics movement is minimized in order for their calibration to be maintained for a very broad range of temperature. In the course of 2015, it is expected that sensors, scanning systems and new products will enter the market to seal the gap between the long to mid range terrestrial scanners and the close range scanning (Eugene, 2015). Handheld scanners are considered to have a very critical role in forensics; particularly they have an advantage in terms of data quality, higher accuracy and capture speed over the traditional documentation methods. Cost is considered to be a factor, but the ease of use as well as integration to other workflows and packages is taken to be more important (Eugene, 2015). Research Method In the study to be undertaken, an alternative solution related to the capturing of the three dimension prints problem at the crime scene is proposed. This will be compared with the solutions that exist that include three dimension scanning and casting where, their drawbacks and advantages will be considered. The solution will include a pipeline where, the three dimension reconstruction will be generated using digital photographs that will be taken at the crime scene from the footwear print. There will be a reconstruction of a complete three dimension model from image collection that has been taken at various viewpoints. The first step will involve recovering camera parameters as well as three dimension locations in each image with the use of Blunder (Snavely, Seitz and Szelisk, 2006) where, it is a proposed method meant for performing structure from motion (SfM) on image collection that is unordered. The second step will involve generating a dense point cloud with the use of PMVS (Furukawa and Ponce, 2009), and it is known to be a patch-based multi-view stereo method. The final step will involve reconstructing the surface with the use of a FARO "Freestyle" 3D Scanner. The experiment will involve the comparison of the FARO "Freestyle" 3D Scanner with other methods used in practice. Three dimension scanning, based on the several footwear prints scanning in sand. A sandbox will be set up with a three dimension scanner attached. In this case, the footwear marks will be obtained and it will have four various types of shoes. Two equipment will be used in the experiment and it includes three dimension laser scanner and Digital camera. The scanner will first be set up at a distance (31.5”) from the sand in the box. Each footwear print will be scanned and then photographs will be taken rotating around it at about equal sized angles. For the comparison purposes, the four shoe soles that correspond to the print will be scanned. The images of each shoe print, twelve in number will be used as input to produce a three dimension reconstructions. The produced three dimension models and the scanned shoe prints will be registered with the corresponding shoe soles using Meshlab (Bodziak, 2000).The Haursdoff distance that is computed by metro tool (Cignon, Rocchini and Scopigno,1998) will be adopted in order to evaluate the produced three dimension model. The distance map dg representing the Haursdoff distance for the shoe print scan vertices will be considered as ground-truth in relation to the shoe sole scan vertices. The dg is considered to be subjective to the shoe sole’s curvature when the shoe is not employed and as a result not flat in relation to the ground. In the same manner, the distance map dm that will be between the produced model and the shoe sole scan will be computed. The value (dm-dg) is considered to measure the quality of the model generated compared to the three dimensions scanning, where, dm and dg represents the averages of the distances in dg and dm respectively. The analysis of the result will be able to determine if the method, FARO "Freestyle" 3D Scanner is able to capture comparable amount of details and the same fact can be asserted by analyzing the value (dm-dg) computed for every shoe print(table below) The template showing the evaluated 3D model for each shoe print using Haursdoff distance Shoe print number dg dm (dm − dg) 1 2 3 Conclusions and recommendations The proposed study will present the method of recovering foot wear impression evidence from the crime scenes. This will be on the basis of the technique referred to as 3D laser scanning of footwear impression evidence with the use of a FARO "Freestyle" 3D Scanner. This technique is believed to have never been used based on such kind of settings in the literature. Therefore, the results obtained will be discussed extensively based on the reconstructed three dimensional models as well as the drawbacks and the advantages of these models. The analysis of the result will be able to determine if the method, FARO "Freestyle" 3D Scanner is able to capture comparable amount of details References Bodziak, W. (2000) Footwear impression evidence: detection,recovery, and examination. USA: CRC Press, 2nd ed. Cignon,i P., Rocchini C., and Scopigno, R.(1998). “Metro: measuring error on simplified surfaces,” in Computer Graphics Forum, vol. 17, pp. 167–174 DeLaurentis P. (2009) “3D Scanning: A New Tool for Cracking Tough Cases,” Forensic Magazine, vol. 6, no. 1, pp. 37–40 Eugene Liscio, P. (2015). Review of Handheld Scanners from FARO, DotProduct, and More. A New Perspective: Guest Blogs. Retrieved from http://www.sparpointgroup.com/guest-blogs/vol13no4-review-of-handheld-scanners Furukawa and Ponce J.( 2009). “Accurate, dense, and robust multi-view stereopsis,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 32, no. 8 Gamage R. EJoshi., A., Zheng J., and Tuceryan M.(2013). A 3d impression acquisition system for forensic applications. In X. Jiang, O. Bellon, D. Goldgof, and T. Oishi, editors, Advances in Depth Image Analysis and Applications, volume 7854 of Lecture Notes in Computer Science, pages 9{20. Springer Berlin Heidelberg, Hess K., and. Orthmann, C. (2010). Criminal Investigation. USA: Delmar Cengage Learning, 9th ed. Houck M. and Siegel J. (2010). Fundamentals of forensic science. Academic Press, 2nd edition, Liscio E., (2009). “A Primer on 3D Scanning in Forensics,” Forensic Magazine,. McDonald P. (1992). Tire imprint evidence. CRC Press, Snavely N., Seitz S., and Szelisk R. (2006). “Photo tourism: exploring photo collections in 3D,” in ACM Transactions on Graphics, vol. 25, pp. 835–846,. Read More