Data imbalance is one of the problems that we face when applying machine learning to real-world problems, especially in image classification. With all the improvements in machine learning, especially deep learning, research in this area is drawing more attention from academics and even industry. To address this imbalanced data problem, we adopt a hybrid (algorithm and data) approach that consists of data manipulation and weighted loss function in this paper. We propose Ripple-SMOTE as a novel oversampling method to generate synthetic data for preprocessing. A deep neural network and the weighted loss function is applied so it will not treat all classes equally. We also use a pre-trained model and fine tune it to improve the classification accuracy. In this paper, we report the evaluation results using imbalanced data sets based on MNIST, CUReT texture set, and Malware data set, and show that our approach significantly improves the performance in imbalanced data cases and outperforms the conventional approaches, especially in handling minority classes.

No abstract

No abstract

This paper presents our research on location estimation based on the identity of floor surface patterns, which we call "floor fingerprints," from a photographic image of a floor taken with a hand-held smartphone. Because floor textures generally appear to lack sufficient features, it may seem difficult using general feature detection algorithms to find matching pairs of features in two corresponding floor images taken at an identical location but from different orientations of the camera and under different lighting conditions. We demonstrate, however, that use of a preprocessing image filter, involving gravity-rectified image adjustment against perspective distortion and enhancement of local image features, provides well-aligned detail sufficient to allow detection of paired features of floor textures. Although the enhancement filter reveals many noisy pairs in local feature detection, we show that it is possible to choose a valid imageto-image correspondence efficiently using our newly proposed B-ORB feature detector and RANSAC. Since matching a query floor image with large-scale floor images stored at the server requires a large amount of processing resources, we utilize GPGPU for the feature detection and matching. This paper proves the feasibility and efficiency of the proposed approach, based on our experimental results concerning the accuracy and processing time, and discusses possible solutions to a wide range of real-world indoor location applications.

No abstract