Image Classification Algorithm Based on Improved AlexNet
Shaojuan Li, Lizhi Wang, Jia Li, Yuan Yao
Abstract
Abstract Aiming at the problems that the traditional CNN has many parameters and a large proportion of fully connected parameters, a image classification method is proposed, which based on improved AlexNet. This method adds deconvolution layer to traditional AlexNet and classifies the images by full connection layer. Using Cifar-10 data set to test the classification algorithm. The results indicate that the method not only reduces the number of parameters and parameters proportion of the full connection layer, but also improves the classification accuracy compared with AlexNet.
Topics & Concepts
Computer scienceDeconvolutionPattern recognition (psychology)Image (mathematics)Layer (electronics)Set (abstract data type)AlgorithmConnection (principal bundle)Artificial intelligenceContextual image classificationMathematicsChemistryProgramming languageOrganic chemistryGeometryAdvanced Neural Network ApplicationsImage Processing and 3D ReconstructionAnomaly Detection Techniques and Applications