In recent years, complex machine learning and deep learning algorithms have been used in prognostics and health management applications to monitor the state of industrial systems in real-time and predict failures in advance. However, the performance of these algorithms depends on the size of the collected labeled datasets. This study proposes a semi-weakly supervised learning method that creates label functions using a small amount of data and, consequently, combines weak supervision and semi-supervision methods to expand the labeled dataset for training. The proposed model training method creates labels automatically in the life prediction dataset of rolling element bearings, which were manually labeled by expert groups. The performance of the deep neural network is continuously improved through experiments, and a performance improvement similar to that of manually labeled training data is presented. Future research applications include meta-learner experiments (e.g., use of pre-trained word embedding or other model architectures) and more appropriate selection methods and definitions for the basic learner.

Auto Labeling Methods Developed Through Semi-Weakly Supervised Learning in Prognostics and Health Management Applications for Rolling Ball Bearing

The surveillance videos taken from dynamic cam-eras are susceptible to multiple security threats like replay attacks, man-in-the-middle attacks, pixel correlation attacks etc. Using unsupervised learning, it is a challenge to detect objects in such surveillance videos, as fixed objects may appear to be in motion alongside the actual moving objects. But despite this challenge, the unsupervised learning techniques are efficient as they save object labelling and model training time, which is usually a case with supervised learning models. This paper proposes an effective computer vision-based object identification algorithm that can detect and separate stationary objects from moving objects in such videos. The proposed Advanced Flow Of Motion (AFOM) algorithm takes advantage of motion estimation between two consecutive frames and induces the estimated motion back to the frame to provide an improved detection on the dynamic camera videos. The comparative analysis demonstrates that the proposed AFOM outperforms a traditional dense optical flow (DOF) algorithm with an average increased difference of 56 % in accuracy, 61 % in precision, and 73 % in pixel space ratio (PSR), and with minimal higher object detection timing.

AFOM: Advanced Flow of Motion Detection Algorithm for Dynamic Camera Videos

UDF-GAN: Unsupervised dense optical-flow estimation using cycle Generative Adversarial Networks

There is a growing consensus in computer vision that symmetric optical flow estimation constitutes a better model than a generic asymmetric one for its independence of the selection of source/target image. Yet, convolutional neural networks (CNNs), that are considered the de facto standard vision model, deal with the asymmetric case only in most cutting-edge CNNs-based optical flow techniques. We bridge this gap by introducing a novel model named SDOF-GAN: symmetric dense optical flow with generative adversarial networks (GANs). SDOF-GAN realizes a consistency between the forward mapping (source-to-target) and the backward one (target-to-source) by ensuring that they are inverse of each other with an inverse network. In addition, SDOF-GAN leverages a GAN model for which the generator estimates symmetric optical flow fields while the discriminator differentiates the “real” ground-truth flow field from a “fake” estimation by assessing the flow warping error. Finally, SDOF-GAN is trained in a semi-supervised fashion to enable both the precious labeled data and large amounts of unlabeled data to be fully-exploited. We demonstrate significant performance benefits of SDOF-GAN on five publicly-available datasets in contrast to several representative state-of-the-art models for optical flow estimation.

Yunshuai Yang

Papers

SDOF-GAN: Symmetric Dense Optical Flow Estimation With Generative Adversarial Networks