Generative Adversarial Networks (GANs) can be used for data augmentation in order to improve the outcome and performance of machine learning models for automatic information retrieval. We looked into the challenge faced with limited blurry and distorted digit images from expiry dates datasets, which is required to improve digit recognition tasks on medicine, consumables, cosmetic products and tube-type ointments. For our dataset, Wasserstein GAN with a gradient norm penalty (WGAN-GP) was effective for data augmentation among the state-of-the-art GANs by visible inspection and Fréchet Inception Distance (FID) value comparison.

Various methods of personal identification and authentication using biometric data have been developed. However, there are downsides to these techniques due to the risk of data leakage during the authentication process and the data acquisition environment required for recognition. We propose a scheme to identify the residents in the daily living space through pairs of real and thermal images taken from Internet of Things (IoT) cameras monitoring the space.

Accident cases are also increasing in proportion to the recent increase in the number of users of shared e-scooters. In order to reduce the accidents, the shared e-scooter platform applies the same maximum speed. We propose an effective maximum speed control to increase safety in e-scooter driving on straight and curved roads according to the boarding weight using a pressure sensor.

Abnormal heat generation problems in electronic devices are increasing. In order to prevent fires caused by abnormal overheating, we propose a fire prevention system using thermal images and a LSTM model. Thermal image, LSTM, Fire prevention, Abnormal heat generation