Classification of nematode image stacks by an information fusion based multilinear approach

In this letter, we present to use an information fusion based multilinear analysis approach to classify multi-focal image stacks. First, image fusion techniques such as the nonsubsampled contourlet transform sparse representation (NSCTSR) are used to combine relevant information of multi-focal images within a given image stack into a single image, which is more informative and complete than any single image in the given image stack. Second, multi-focal images within a stack are fused along 3 orthogonal directions, and multiple features extracted from the fused images along different directions are combined by using canonical correlation analysis (CCA). Finally, because multi-focal image stacks represent the effect of different factors - texture, shape, different instances within the same class and different classes of the objects, we embed the information fusion methods within a multilinear analysis (MA) framework to propose an information fusion based multilinear classifier. The experimental results demonstrated that the information fusion based multilinear classifier can reach a higher classification rate (96.6%) than the previous multilinear based approach (86.4%), even we only use the texture feature instead of the combination of texture and shape features as in the previous work.