"... In PAGE 6: ... The experimental results show that palmprint identification based on feature extraction in the frequency domain is effective in terms of accuracy and efficiency. Table4 shows the results of testing. Table 4.... ..."

"... In PAGE 2: ... We have clas- sified each work according to the primary representation used in the recognition algorithm, much in the spirit of [7]. Table3 summarizes the recent work on 3D and 2D+3D face recognition. 2.... ..."

"... In PAGE 2: ... We have clas- sified each work according to the primary representation used in the recognition algorithm, much in the spirit of [7]. Table3 summarizes the recent work on 3D and 2D+3D face recognition. 2.... ..."

"... In PAGE 10: ... The experimental results show that palmprint identification based on feature extraction in the frequency domain is effective in terms of accuracy and efficiency. Table3 shows the results of testing. Table 3 - The results of the palmprint performance testing.... ..."

"... In PAGE 2: ... A detailed de- scription of these approaches is beyond the scope of this paper. However a summary of these approaches with the typical references can be seen in Table1 . Researchers have shown promising results on inked images [2], images ac- quired directly from the scanner [3] and images acquired from digital camera [4] using constrained pegged setup.... In PAGE 2: ... Therefore this paper utilizes such images to inves- tigate the performance improvement. A summary of prior work in Table1 shows that there has not been any attempt to investigate the palmprint authentication using its multiple representations. Several matching score level fusion strategies for com- bining various biometric modalities have been presented in the literature.... ..."

"... In PAGE 4: ... In total, we used 218 range images of 109 subjects for training, where each individual has two samples, and 758 range images for testing, where individuals have different numbers of samples ranging from two to twelve. Table4 shows the results of our approach and the other well known face recognition approaches. Our algorithm is found to be superior also in the 3D domain.... ..."

"... In PAGE 3: ... However, 3D information inherently makes a face recognition system more robust to pose and expression variation. Approach 3 attempts to store face information as a generic 3D model of the face and then mach this model to 2D images (lower left box in Table1 ). This approach is better because it does not require any special hardware for acquiring the face image.... In PAGE 3: ... However, because we have access to a full 3D scanner we have also developed a full 3D face recognition system in the fourth approach (lower right box in Table 1). Note that we did not work on last option (upper right box in Table1 ) where the testing images are 3D faces and the training images are 2D. Table 1.... ..."

"... In PAGE 7: ... Table 2 shows the recognition rate when the input was the training set. Table3 shows the correspond- ing results when the input used the unseen test set. Some interesting observations are in order.... In PAGE 7: ... Some interesting observations are in order. First, Table 2(a) and Table3 (a) show that the recognition rate for the DNN drops sharply on the unseen test input, verifying that the DNN indeed lacks Fig. 6 Comparisons of input potentials Ii when the input test se- quences are CABDABE and CABAABE.... In PAGE 8: ...8% 100% mean 97.1% Table 2(d) Confusion table of Hybrid System Test Data Speech Model B D G B8000 D08 G08 Recognition Rate 100% 100% 100% mean 100% Table3 (a) Confusion table of DNN (Testing) Test Data Speech Model B D G B2027 D219 G3754 Recognition Rate 28.5% 30% 77% mean 45.... In PAGE 8: ...5% 30% 77% mean 45.17% Table3 (b) Confusion table of STNN (Testing) Test Data Speech Model B D G B52528 D1374 G583 Recognition Rate 74% 81.4% 54.... In PAGE 8: ...4% mean 85.7% Table3 (c) Confusion table of HMM (Testing) Test Data Speech Model B D G B5984 D10463 G16 Recognition Rate 84.3% 65.... ..."