MOMENTSNET A SIMPLE LEARNING-FREE METHOD FOR BINARY IMAGE RECOGNITION

Jiasong Wu (1, 2, 3) , Shijie Qiu (3, 1) , Youyong Kong (3, 1) , Yang Chen (1, 3) , Lotfi Senhadji (2, 3) , Huazhong Shu (1, 3)

1 LIST - Laboratory of Image Science and Technology [Nanjing]
2 LTSI - Laboratoire Traitement du Signal et de l'Image
3 CRIBS - Centre de Recherche en Information Biomédicale sino-français

In this paper, we propose a new simple and learning-free deep learning network named MomentsNet, whose convolution layer, nonlinear processing layer and pooling layer are constructed by Moments kernels, binary hashing and block-wise histogram, respectively. Twelve typical moments (including geometrical moment, Zernike moment, Tchebichef moment, etc.) are used to construct the MomentsNet whose recognition performance for binary image is studied. The results reveal that MomentsNet has better recognition performance than its corresponding moments in almost all cases and ZernikeNet achieves the best recognition performance among MomentsNet constructed by twelve moments. ZernikeNet also shows better recognition performance on a binary image database than that of PCANet, which is a learning-based deep learning network.

Mots clés: Deep learning; convolutional neural network; PCANet; MomentsNet; Zernike moment; feature extraction; image recognition; learning (artificial intelligence); neural nets; Deep learning architecture

Domaines: Traitement du signal et de l'image [eess.SP]; Ingénierie biomédicale

Jiasong Wu, Shijie Qiu, Youyong Kong, Yang Chen, Lotfi Senhadji, et al.. MOMENTSNET A SIMPLE LEARNING-FREE METHOD FOR BINARY IMAGE RECOGNITION. 24th IEEE International Conference on Image Processing (ICIP), Sep 2017, Beijing, China. pp.2667-2671, ⟨10.1109/ICIP.2017.8296766⟩. ⟨hal-01807157⟩