講座題目：LocalN-ary Pattern and Its Extension for Texture Classification
主講人：何祥健（Universityof Technology, Sydney）
講座時間：2015年4月14日10:00-11:30
講座地點：經管樓306
?
講座內容簡介：
Textureimage classification is important in computer vision research. In order toeffectively capture texture patterns, a distinctive feature such as localbinary pattern (LBP) is needed. LBP is robust against monotonic and gray- scalevariations and it is fast to compute. Its robustness and speed advantage hasmade it popular in various texture analysis applications. However, LBP issensitive to noise, particularly smooth weak illumination gradients in near-uniform regions. In order to mitigate the effect of noise and increasedistinctiveness, a local ternary pattern (LTP) isproposed. Compared to the binary coding LBP, LTP adopts ternary coding. As aresult, LTP can better tolerate noise and is significantly more distinctive.These advantages of LTP effectively improve its classification accuracy.However, the potential of ternary coding is not fully explored in LTP becausethe ternary pattern is spitted into a pair of binary patterns. In our work, inorder to fully explore the distinctiveness in the local pattern, the featureextraction process is formulated as an integer decomposition problem, which isa generalized version of the Bachet de Meziriac Weight Problem (BMWP).Following this generalization, a local n-ary pattern (LNP) is proposed, forwhich the LBP is a special case parametrized under n = 2. The LTP isnot a special case of the LNP. Both LBP and LTP are used as benchmark methodsto evaluate the LNP’s performances due to their well-recognized success. Inaddition, rotation invariant and uniform LNP is also proposed and compared torotation invariant and uniform LBP. The proposed LNP achieves significantlyimproved texture classification accuracy when compared to LBP and alsodemonstrates considerable improvement over LTP.
?
主講人簡介：
ProfessorXiangjian He, as a Chief Investigator has received various research grantsincluding four national Research Grants awarded by Australian Research Council(ARC).
Heis the Director of Computer Vision and Recognition Laboratory at the Global BigData Technologies Centre (GBDTC) at the University of Technology, Sydney (UTS).
Heis an IEEE Senior Member and an IEEE Signal Processing Society StudentCommittee member. He has been awarded 'Internationally Registered TechnologySpecialist' by International Technology Institute (ITI). He has been carryingout research mainly in the areas of image processing, network security, patternrecognition and computer vision in the previous years. He is a leadingresearcher for image processing based on hexagonal structure. He has playedvarious chair roles in many international conferences such as ACM MM, MMM, IEEECIT, IEEE AVSS, TrustCom and ICARCV.
Inrecent years, he has many high to top quality publications in IEEE Transactionsjournals such as IEEE Transactions on Computers, IEEE Transactions on Paralleland Distributed Systems, IEEE Transactions on Circuits and Systems for VideoTechnology, IEEE Transactions on Reliability, IEEE Transactions on ConsumerElectronics, and in Elsevier’s journals such as Signal Processing, Neuro-computing,Future Generation Computer Systems, Computer Networks, Computer and SystemSciences, Network and Computer Applications. He has also had papers publishedin premier international conferences and workshops such as CVPR, ECCV, ACM MMand WACV. His papers have been cited thousands of times.
Hehas recently been a guest editor for various international journals such asJournal of Computer Networks and Computer Applications (Elsevier) and SignalProcessing (Elsevier). He has also been in the editorial boards of variousinternational journals.
Hehas been a supervisor of postdoctoral research fellows and PhD students.
Since1985, he has been an academic, a visiting professor, an adjunct professor, apostdoctoral researcher or a senior researcher in variousuniversities/institutions including Xiamen University, China, University of NewEngland, Australia, University of Georgia, USA, Electronic andTelecommunication Research Institute (ETRI) of Korea, University of Aizu,Japan, HongKong Polytechnic University, and Macau University.
?
