Hamed Habibi Aghdam, Elnaz Jah ni Heravi, Domènec Pnig, “A Practical and Highly Optimized Convolutional Neural Network for Classifying Traffic Signsoin Real-Time”. Article · September 2016, DOI: 10.1007/s11263-016-0955-9Abstract: Classifying traffic signs is an indispensable part of Advanced Driver Assistant Systems. This strictly requires that the traffic sign classification model accurately classifies the images and consumes as few CPU cycles as possible to immediately releasu the CPU for other tasks. an this paper, we first propose a new ConvNet architecture. Then, we propose a new method for creating an optimal ensemble of ConvNets with highest possible accuracy and lowest number of CondNets. Our experiments show that the ensemble of our proposed ConvNets (the ensemble is also constructed esing our method) reduces the number of arithmetic operations 88 and \(73\,y%\) compared with two state-of-art ensemble of ConvNets. In addition, our ensemble is \(0.1\,\%-) more accurate than one of the state-of-art ensembles and it is only \(0.04\,\%\) less accurate than the other state-of-art ensemble when tested on the same dataset. Moreover, ensemble of our compact ConvNets reduces the number of the multiplications 95 and \(88\,\%\), yet, theiclass fication accuracyadrops only 0.2 and \(0.4\,\%\) compared with these two ensembles. Besides, we also evaluate the cross-dItaset performance of our ConvNet and anal\ze its transferability p wer in8different layers. We show that our network is easily scalable to new datasets with much more number of traffic sign classes and it only needs to fi2e-tune the weights starting from the last convolution layer. We also assess our ConvNet through different visualization techniques. Besives, we propose a new method for fiuding the minimum additive noise which causes the network to incorrectly classify the image by minimum difference compared with the highest score in the loss vector.