Research On Automated3-D Firearms Marks Identification System And Its’ Key Technology

Ballistics comparison based on Firearms marks’is an important research areaabout the technology of the Criminal Investigation, making use of bullet trailinspection theory to judge and find out the emission guns by identifying the bullettrail, and provide important evidence of gun-related cases for the investigationprocess. At the current of our country, using microscope to observe and compare isthe major method, but this inefficient and low accuracy rate approach can’t meet thebullet identification efficiency and accuracy needs. With the development ofcomputer technology, the automatic bullet trail recognition technology is an effectiveway to identify the bullet trail. Bullet trail recognition technology is which make useof inherent trail features of bullet to identify the emission guns by digital imageprocessing and pattern recognition theory. Since the importance and practicality ofbullet trail recognition technology, domestic and foreign scholars have attached moreimportance to it.Bullet trail comparison technology has developed in recent years in our country,compared to the developed countries, our research is all based on the twodimensional gray-scale image, due to the two dimensional gray-scale images areinfluenced hardly by light intensity, angle and other environmental factors in theprocess of data acquisition, and the two dimensional image can’t reflect thetopography information of bullet trail, so, it can’t guarantee the reliability of the identification result and the final discrimination of two bullets need to identifymanually, the identification efficiency doesn’t improve essentially. And through therelated fields study of other countries, the depth of the surface texture information ofbullet is an important object of bullet trail identification research, so, according tothe bullet trail comparison technical requirements, that ensure high accuracy whileavoiding to damage the texture trail of bullets again. In this dissertation, athree-dimensional non-contact optical measurement system is adopted to acquire theinformation of bullets, and it can simultaneously obtains high precisionthree-dimensional topography data and two-dimensional gray-scale images, andintroduces the principle, structure, and the calibration process of the device andbullet ROI regional acquisition process. Bullet trail recognition is a high precisiontrace recognition technology, and it is necessary to study on it deeply, but the bullettrail identification and matching technology in our country is still in its infancy, itstheory and application should be research further and improvement. Therefore, thekey technical issues of bullet identification are studied deeply in this dissertationbased on high precision bullet trail images, and have proposed effective algorithmsabout2D+3D image fusion, feature extraction of bullet/cartridge, bullet trailidentification and bullet trail images retrieval based on multi-feature fusion and otherkey technologies.In this paper，we analyze the authenticity of the three dimensional firearmmarks topography data, and taking into account the two-dimensional image of thebullet also contains a mass of important texture information, combined with theactual needs, an image fusion strategy which fuses2D ballistics image based onmulti-scale Contourlet transform and3D ballistics Gaussian curvature image isproposed to improve the recognition rate, and ensure the objectivity and reliability.On this basis, for the character of ballistics, we proposed main traces extractionalgorithm of bus based on AFRAT and texture feature extraction algorithm based ontwo-dimensional Log-Gabor and Wavelet transform. Similarly, separate the trail offiring pin and breech face marks according to the characteristics of cartridge trail, and discuss the shape feature extraction of firing pin trail and texture featureextraction of breech face marks based on PCNN algorithm. After extract the featuresof ballistics independently, for issue of low recognition rate of traditional signalfeature, propose a recognition retrieval method based on multi-features fusionthrough the study of traditional feature matching algorithms, In this paper,wepropose a retrieval method based on multi feature fusion, independentfeatures formed by multiple sets of features, using the matching level fusiondecision, based on multi feature recognition bullet image content retrievalmodel, proved by experiment, achieved satisfactory identification results. Based onthe ballistics recognition model, we research the multistage retrieval method basedon the strategy of the bullet image, and the bullet image identified according tothe similarity in descending order, feedback to the user.Finally, we designvisualization experiment platform, and through the analysis and discussions ofmultiple experiments, verified the feasibility and effectiveness of the algorithm andexperimental platform systems.