An Approach Combined The Faster RCNN And MobileNet For Logo Detection

Theory of digital image processing is one of the main sections in computer science.As part of it,algorithms for object detection,their analysis,and computer vision have been developed.A man has long been trying to teach a computer to understand what is on the image.What seems to us to be a completely trivial task,the developed algorithms are still not capable.For the beginning it is necessary to solve simpler problems,i.e.,it is necessary to learn how to find the key objects of the real world in the image and to understand what these objects are.In any case,images on social media provide a useful source of information for market research.The reliable detection of logo forms an important building block for any further analysis of these brand interactions.Although deep learning object detection tools such as Faster Recurrent Convolution Neural Network has demonstrated good performances in object detection,they also have a limited success rate for some applications.It is due to the lack of refinedness of feature maps for accurate localization,the insensitivity for small scal e objects and fixed-window feature extraction in Region Proposal Network.This paper performed a meticulous examination of both the proposal and the classification process by evaluating the adequacy of feature representations from different stages of the feature sequencing.For trainable object proposals we look at Region Proposal Networks which we analyze in detail,both theoretically and in practice and notice some fundamental shortcomings for detecting logos.The approach presented an approach to improve the Regional Proposal Network by appropriate anchors selection,and proposed a modification by combining Faster R-CNN and MobileNet which influences higher-resolution feature maps for mobile devices.The results demonstrate that Faster R-CNN architecture with MobileNet has the best detection accuracy.The experiment result showed that we managed to achieve a final accuracy of 92.4% on a NVIDIA GeForce Gtx 1070 compare to previous work that achieved 90.8% of accuracy and found that our models performed well at the detection,with very low false positive rates possible for a fairly reasonably.