27 January 2022 in IEEE Transactions on Components, Packaging and Manufacturing Technology
In the mass production of electronic products, in-circuit-test (ICT) and printed circuit board assembly (PCBA) quality tests are performed. ICT measures resistance values and capacitance, but not only does it require the use of a fixture that is expensive and requires frequent replacement but also the fixture’s needles may cause PCBA defects. To overcome these limitations, various studies tried to replace ICT using visual inspection methods; however, visual inspection methods cannot be applied to chip resistors and chip capacitors that do not have externally visible characteristics. In this article, we propose a contactless inspection method that can detect PCBA defects without the use of the fixture and ICT by using the comparison of thermal images and deep learning (DL) analysis. We review the existing contactless inspection methods and compare them with our proposed thermal image analysis method. We analyzed thermal images by applying a structural similarity index map as a rule-based object detection method, and we used convolutional neural networks (CNNs), regions with CNN features, and an autoencoder as DL analysis methods. As a result, we achieved highly accurate defective component detection and location in real time.
11 December 2024 in IEEE Transactions on Intelligent Transportation Systems
Sound can be utilized to gather information about vehicles approaching a Non-Line-of-Sight (NLoS) region that remains hidden from Line-of-Sight (LoS) sensors due to its reflective and diffractive characteristics, like a radar. However, due to the inability to determine the location of NLoS vehicles in previous studies, it has not been possible to construct a sound-based active emergency braking system. This paper introduces a novel approach for localization of vehicles approaching in NLoS regions through sound. Specifically, a new particle filter method incorporating Acoustic-Spatial Pseudo-Likelihood (ASPLE) has been proposed to track objects using both acoustic and spatial information from the ego vehicle. Also, the Acoustic Recognition based Invisible-target Localization (ARIL) dataset, which is the firstly providing the location of the NLoS vehicle as ground truth using Bird’s Eye View camera, is proposed. The proposed method is validated using two datasets: the ARIL dataset and the Occluded Vehicle Acoustic Detection Dataset (OVAD) dataset. The proposed method exhibited remarkable performance in localizing NLoS targets in both datasets, predicting the location of the vehicle in the NLoS region. Lastly, the analysis of how the reflection of sound affects to the proposed method, highlighting variations based on the spatial situations, and demonstrate the empirical convergence of the method is described. Our code and dataset is available at https://github.com/mingujeon/NLoSVehicleLocalization.
17 February 2023 in IEEE Internet of Things Journal
Accurately estimating the value of an equipment is a significant challenge in the industrial environment. Conventional methods mainly considered discounting the value over time, but they are limited in that they do not consider the status of individual equipment. Recent developments in Industrial Internet of Things (IIoT) and AI technologies have opened up the possibility of real-time remote monitoring on the status of machinery, thus providing an opportunity to more accurately estimate the value of machine equipments. In this study, we designed a sensor that can acquire the vibration and magnetic field data of an equipment, with which we proposed a 1-D convolutional neural network that can classify the status of machinery based on the data obtained by the designed sensor. In addition, based on the results of the classification model, the cumulative fatigue of equipment was predicted using Pålmgren–Miner’s linear damage rule, with which we proposed a model for estimating the value of the movable property based on the cumulative fatigue.
5 September 2024 in IEEE Transactions on Components, Packaging and Manufacturing Technology
With advancements in precise semiconductor manufacturing processes, a new category of anomalies has increasingly emerged. However, due to the probability of an abnormal occurrence during the semiconductor bonding process being less than 1 in 10 million, conventional statistical methods and supervised learning-based neural networks face significant limitations in detecting these anomalies. To address this, several data augmentation techniques have been proposed, yet they fail to ensure the similarity of the augmented time series data. In response, this study proposes a time series data augmentation method using digital twins to address the extreme class imbalance problem, and presents a pipeline that incorporates this method with an autoencoder-based anomaly detection approach. A robotic arm for the bonding process of non-ductile materials was designed to closely mimic the actual process, reflecting the physical properties of the robotic arm, non-ductile materials, and particles. The effectiveness of this approach was validated by applying the optimized anomaly score threshold derived from the augmented data to detect anomalies in the actual manufacturing process. This study not only presents an anomaly detection method capable of selecting the most representative patterns from numerous normal samples for comparison with abnormal data but also offers valuable insights into addressing the challenge of detecting extremely rare anomalies.