Search

From traditionally handmade items to the ability of people to use machines to process and even to human-robot collaboration, there are many risks. Traditional manual lathes and milling machines, sophisticated robotic arms, and computer numerical control (CNC) operations are quite dangerous. To ensure the safety of workers in automated factories, a novel and efficient warning-range algorithm is proposed to determine whether a person is in the warning range, introducing YOLOv4 tiny-object detection algorithms to improve the accuracy of determining objects. The results are displayed on a stack light and sent through an M-JPEG streaming server so that the detected image can be displayed through the browser. According to the experimental results of this system installed on a robotic arm workstation, it is proved that it can ensure recognition reaches 97%. When a person enters the dangerous range of the working robotic arm, the arm can be stopped within about 50 ms, which will effectively improve the safety of its use.

Anomalies are a set of samples that do not follow the normal behavior of the majority of data. In an industrial dataset, anomalies appear in a very small number of samples. Currently, deep learning-based models have achieved important advances in image anomaly detection. However, with general models, real-world application data consisting of non-ideal images, also known as poison images, become a challenge. When the work environment is not conducive to consistently acquiring a good or ideal sample, an additional adaptive learning model is needed. In this work, we design a potential methodology to tackle poison or non-ideal images that commonly appear in industrial production lines by enhancing the existing training data. We propose Hierarchical Image Transformation and Multi-level Features (HIT-MiLF) modules for an anomaly detection network to adapt to perturbances from novelties in testing images. This approach provides a hierarchical process for image transformation during pre-processing and explores the most efficient layer of extracted features from a CNN backbone. The model generates new transformations of training samples that simulate the non-ideal condition and learn the normality in high-dimensional features before applying a Gaussian mixture model to detect the anomalies from new data that it has never seen before. Our experimental results show that hierarchical transformation and multi-level feature exploration improve the baseline performance on industrial metal datasets.