flâneur

Image Classification using SVM and CNN | IEEE Conference Publication | IEEE Xplore

ieeexplore.ieee.org · 815 words · saved by 1 readers

To develop an understanding of image classification from the ground-up, i.e., start with the traditional machine learning techniques first and then move onto deep learning techniques. To research the underlying principles and techniques of image classification and to highlight the potential of deep learning techniques and illustrate their importance over the more traditional machine learning techniques. As mentioned before, we began with the more traditional algorithms for classification. After having gone through most of them, we chose to implement SVM. Although it made more sense to implement Neural Networks (NN), the more modern approach, we felt that SVM would be a better starting point and we weren’t wrong. Now before we begin with what SVM is and how we implemented it, we think we require to explain why we thought that NN should’ve been the way to go. Accuracy of Image classification using NN far exceeds any machine learning algorithm and even other deep learning algorithms. The

Download PDF Download References Request Permissions Save to Alerts Abstract: On the surface, teaching a computer to do something like image classification seemed very intriguing to us. Moreover, there are countless real-world applications of this ...Show More Metadata Abstract: On the surface, teaching a computer to do something like image classification seemed very intriguing to us. Moreover, there are countless real-world applications of this concept. It is in light of these reasons that we decided to work on Image Classification. Thankfully though, this topic has been…

saved by

related reading