Logo Recognition with the Use of Deep Convolutional Neural Networks
Ahmed A. Alsheikhy, Yahia Said, Mohammad Barr
Abstract
Automatic logo recognition is gaining importance due to the increasing number of its applications. Unlike other object recognition tasks, logo recognition is more challenging because of the limited amount of the available original data. In this paper, the transfer leaning technique was applied to a Deep Convolutional Neural Network model to guarantee logo recognition using a small computational overhead. The proposed method was based on the Densely Connected Convolutional Networks (DenseNet). The experimental results show that for the FlickrLogos-32 logo recognition dataset, our proposed method performs comparably with state-of-the-art methods while using fewer parameters.
Topics & Concepts
Convolutional neural networkLogo (programming language)Computer scienceArtificial intelligenceCognitive neuroscience of visual object recognitionPattern recognition (psychology)Deep learningTransfer of learningOverhead (engineering)Object (grammar)Artificial neural networkNeocognitronSpeech recognitionTime delay neural networkOperating systemProgramming languageAdvanced Image and Video Retrieval TechniquesHandwritten Text Recognition TechniquesMultimodal Machine Learning Applications