Jinfei Shi

Prof. Jinfei Shi

orcid

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
26 May 2022 in Sensors
Verified
Journal Article
Open access
PDF file
DXDongting Xu
Dongting Xu
ZZZhisheng Zhang
Zhisheng Zhang
JSJinfei Shi
Jinfei Shi
Multiple sensors are often mounted in a complex manufacturing process to detect failures. Due to the high reliability of modern manufacturing processes, failures only happen occasionally. Therefore, data collected in practical manufacturing processes are extremely imbalanced, which often brings about bias of supervised learning models. Data collected by the multiple sensors can be regarded as multivariate time series or multi-sensor stream data. The high dimension of multi-sensor stream data makes building models even more challenging. In this study, a new and easy-to-apply data augmentation approach, namely, imbalanced multi-sensor stream data augmentation (IMSDA), is proposed for imbalanced learning. IMSDA can generate high quality of failure data for all dimensions. The generated data can keep the similar temporal property of the original multivariate time series. Both raw data and generated data are used to train the failure detection models, but the models are tested by the same real dataset. The proposed method is applied to a real-world industry case. Results show that IMSDA can not only obtain good quality failure data to reduce the imbalance level but also significantly improve the performance of supervised failure detection models.

28 March 2022 in Electronics
Verified
Journal Article
Open access
PDF file
DXDongting Xu
Dongting Xu
ZZZhisheng Zhang
Zhisheng Zhang
JSJinfei Shi
Jinfei Shi
Manufacturers are struggling to use data from multiple products production lines to predict rare events. Improving the quality of training data is a common way to improve the performance of algorithms. However, there is little research about how to select training data quantitatively. In this study, a training data selection method is proposed to improve the performance of deep learning models. The proposed method can represent different time length multivariate time series spilt by categorical variables and measure the (dis)similarities by the distance matrix and clustering method. The contributions are: (1) The proposed method can find the changes to the training data caused by categorical variables in a multivariate time series dataset; (2) according to the proposed method, the multivariate time series data from the production line can be clustered into many small training datasets; and (3) same structure but different parameters prediction models are built instead of one model which is different from the traditional way. In practice, the proposed method is applied in a real multiple products production line dataset and the result shows it can not only significantly improve the performance of the reconstruction model but it can also quantitively measure the (dis)similarities of the production behaviors.

14 February 2022 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
CMChao Meng
Chao Meng
JSJinfei Shi
Jinfei Shi
FHFei Hao
Fei Hao
PLPanyu Li
Panyu Li
In the manufacturing process of the wafer, the positioning accuracy of the wafer automatic optical inspection (AOI) system affects the detection of die surface defect directly. Most existing error compensation methods generally analyze specific error or main sources of error and can improve the positioning accuracy in most of the applications. However, to fulfill the requirement of higher positioning accuracy of the wafer AOI system, in this article, we propose an error calibration method based on perspective mapping (ECM-PM), which only requires to focus on the total error of the system, regardless of the multiple error sources of the system. First, the error cause of the wafer AOI system is analyzed and the total error of the system is measured using the designed high-precision glass checkerboard calibration board. Then, the mapping model between motor coordinate system and checkerboard coordinate system is established. In addition, the total error of the system is calibrated based on the perspective mapping principle. Finally, the experimental results show that the proposed method obviously decreased the total error of the system and it is robust to the parameter of different recipe motor coordinates.

17 November 2023 in Measurement Science and Technology
Verified
Journal Article
Closed access
CMChao Meng
Chao Meng
FHFei Hao
Fei Hao
PLPanyu Li
Panyu Li
JSJinfei Shi
Jinfei Shi
Automatic defect detection for glassivation passivation parts (GPP) wafer surface becomes an extremely challenging task, due to the interference of random texture, disturbance of low-contrast pseudo defects within the image, and difference of different brightness between images. In this paper, we propose a novel defect detection scheme for GPP wafer surface with random texture and different brightness. First, an automatic segmentation method center expansion idea-based for the region of interest (ROI) of die image is presented to eliminate the interference from edge background and improve the efficiency of defect detection. Then, a feature point set extraction method based on adaptive gain and error diffusion is proposed so that the defect feature between images with different brightness can be highlighted by adaptive gain, and the candidate defect feature point set under random texture feature is able to be extracted by error diffusion. Finally, the density-based spatial clustering of applications with noise considering grayscale constraint (GC-DBSCAN) is designed to identify the true defect clusters from the candidate defect feature point set and accomplish the detection of various types of surface defects. Experimental results show that the proposed method can completely implement the extraction of crack and dirty defects and eliminate the false detection caused by random texture and different brightness, which is very efficient and superior to other methods.

22 May 2021
Verified
Conference Paper
Open access
CMChao Meng
Chao Meng
JSJinfei Shi
Jinfei Shi
FHFei Hao
Fei Hao
YCYuan Chao
Yuan Chao
To improve the accuracy of kerf angle, an automatic calibration method for kerf angle in wafer automated optical inspection is presented. First, the error model of inspection system is established and system angle deviations are calibrated. Next, normalized positioning-based the kerf edges of interest in multiple images are extracted. Then, the coordinate transformation considering the system angle deviation compensation is performed. Finally, the kerf edge line is fitted based on the least squares method to obtain the kerf angle and the kerf angle can be automatically calibrated by rotating the stage. The experimental results show that the kerf angle obtained is relatively stable by coordinate transformation of multiple images to enhance the information of kerf edge and the accuracy of kerf angle can reach within 0.02 degree. Besides, the kerf angle is more sensitive to the system angle deviation and the result is basically a linear increase.

29 January 2023
Verified
Conference Paper
Closed access
JSJinfei Shi
Jinfei Shi
WJWeijia Jiang
Weijia Jiang
YHYali Han
Yali Han
WZwenliang zhu
wenliang zhu
A multimodal elastic actuator based on the brake and clutch device of worm gear is designed for efficient legged robot locomotion. Several modes of operation can be outputted with the engaging and disengaging of the brake module, including motor drive, series elastic actuation, the energy storage and release. The simulation and performance experiments are applied for the multimodal actuator. The results that the output bandwidth of the system increases gradually with the increase of spring coefficient, and the knee robot leg based on multimodal elastic actuator can realize switchable mode during walking cycle.