An official website of the United States government
The preparation of defect-free wafers serves as a critical stage prior to fabrication of devices or chips as it is not possible to pattern any devices or chips on a defected wafer. Throughout the semiconductor process, various defects are introduced, including random particles that necessitate accurate identification and control. In order to effectively inspect particles on wafers, this study introduces a wafer particle inspection technique that utilizes computer vision based on HSV (hue-saturation-value) color space transformation models to detect and to classify different particles by types. Artificially generated particle images based on their color properties were used to verify HSV color space models of each particle and to demonstrate how the proposed method efficiently classifies particles by their types with minimum crosstalk. A high-resolution microscope consisting of an imaging system, illumination system, and spectrometer was developed for the experimental validation. Micrometer-scale particles of three different types were randomly placed on the wafers, and the images were collected under the exposed white light illumination. The obtained images were analyzed and segmented by particle types based on pre-developed HSV color space models specified for each particle type. By employing the proposed method, the presence of particles on wafers can be accurately detected and classified. It is expected to inspect and classify various wafer particles in the defect binning process.
more »
« less
Hybrid Semiconductor Wafer Inspection Framework via Autonomous Data Annotation
In smart manufacturing, semiconductors play an indispensable role in collecting, processing, and analyzing data, ultimately enabling more agile and productive operations. Given the foundational importance of wafers, the purity of a wafer is essential to maintain the integrity of the overall semiconductor fabrication. This study proposes a novel automated visual inspection (AVI) framework for scrutinizing semiconductor wafers from scratch, capable of identifying defective wafers and pinpointing the location of defects through autonomous data annotation. Initially, this proposed methodology leveraged a texture analysis method known as gray-level co-occurrence matrix (GLCM) that categorized wafer images—captured via a stroboscopic imaging system—into distinct scenarios for high- and low-resolution wafer images. GLCM approaches further allowed for a complete separation of low-resolution wafer images into defective and normal wafer images, as well as the extraction of defect images from defective low-resolution wafer images, which were used for training a convolutional neural network (CNN) model. Consequently, the CNN model excelled in localizing defects on defective low-resolution wafer images, achieving an F1 score—the harmonic mean of precision and recall metrics—exceeding 90.1%. In high-resolution wafer images, a background subtraction technique represented defects as clusters of white points. The quantity of these white points determined the defectiveness and pinpointed locations of defects on high-resolution wafer images. Lastly, the CNN implementation further enhanced performance, robustness, and consistency irrespective of variations in the ratio of white point clusters. This technique demonstrated accuracy in localizing defects on high-resolution wafer images, yielding an F1 score greater than 99.3%.
more »
« less
- PAR ID:
- 10552500
- Publisher / Repository:
- Journal of Manufacturing Science and Engineering
- Date Published:
- Journal Name:
- Journal of Manufacturing Science and Engineering
- Volume:
- 146
- Issue:
- 7
- ISSN:
- 1087-1357
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
null (Ed.)High-resolution optical imaging systems are quickly becoming universal tools to characterize and quantify microbial diversity in marine ecosystems. Automated detection systems such as convolutional neural networks (CNN) are often developed to identify the immense number of images collected. The goal of our study was to develop a CNN to classify phytoplankton images collected with an Imaging FlowCytobot for the Palmer Antarctica Long-Term Ecological Research project. A medium complexity CNN was developed using a subset of manually-identified images, resulting in an overall accuracy, recall, and f1-score of 93.8%, 93.7%, and 93.7%, respectively. The f1-score dropped to 46.5% when tested on a new random subset of 10,269 images, likely due to highly imbalanced class distributions, high intraclass variance, and interclass morphological similarities of cells in naturally occurring phytoplankton assemblages. Our model was then used to predict taxonomic classifications of phytoplankton at Palmer Station, Antarctica over 2017-2018 and 2018-2019 summer field seasons. The CNN was generally able to capture important seasonal dynamics such as the shift from large centric diatoms to small pennate diatoms in both seasons, which is thought to be driven by increases in glacial meltwater from January to March. Moving forward, we hope to further increase the accuracy of our model to better characterize coastal phytoplankton communities threatened by rapidly changing environmental conditions.more » « less
-
Abstract Wafer quality control is one of the important processes to improve the yield rate of semiconductor products. Profile quality and defects in the wafer are two key factors that should be taken into consideration. In this research, we introduce a method that measures the profile of the upper surface and the thickness of the wafer at the same time using an optical fiber cascaded Fabry–Pérot interferometer working at wavelength of 1550 nm. Therefore, the 3D profile of the wafer can be reconstructed directly. Testing results show that both accuracy and precision of the Fabry–Pérot interferometer are within a nanometer scale. Defects, especially those embedded inside the wafer, will be detected by monitoring the leaky field with treating wafers as slab waveguides. With the leaky field detection, defects on the lower surface of the wafer were successfully detected by monitoring the leaky field above the upper surface of the wafer. Compared with traditional methods such as radiographic testing or computed tomography testing, the proposed methods provide a cost-effective alternative for wafer quality evaluation.more » « less
-
Abstract High-resolution optical imaging systems are quickly becoming universal tools to characterize and quantify microbial diversity in marine ecosystems. Automated classification systems such as convolutional neural networks (CNNs) are often developed to identify species within the immense number of images (e.g., millions per month) collected. The goal of our study was to develop a CNN to classify phytoplankton images collected with an Imaging FlowCytobot for the Palmer Antarctica Long-Term Ecological Research project. A relatively small CNN (~2 million parameters) was developed and trained using a subset of manually identified images, resulting in an overall test accuracy, recall, and f1-score of 93.8, 93.7, and 93.7%, respectively, on a balanced dataset. However, the f1-score dropped to 46.5% when tested on a dataset of 10,269 new images drawn from the natural environment without balancing classes. This decrease is likely due to highly imbalanced class distributions dominated by smaller, less differentiable cells, high intraclass variance, and interclass morphological similarities of cells in naturally occurring phytoplankton assemblages. As a case study to illustrate the value of the model, it was used to predict taxonomic classifications (ranging from genus to class) of phytoplankton at Palmer Station, Antarctica, from late austral spring to early autumn in 2017‐2018 and 2018‐2019. The CNN was generally able to identify important seasonal dynamics such as the shift from large centric diatoms to small pennate diatoms in both years, which is thought to be driven by increases in glacial meltwater from January to March. This shift in particle size distribution has significant implications for the ecology and biogeochemistry of these waters. Moving forward, we hope to further increase the accuracy of our model to better characterize coastal phytoplankton communities threatened by rapidly changing environmental conditions.more » « less
-
Additive Manufacturing (AM) allows increased complexity which poses challenges to quality-control (QC) and non-destructive evaluation (NDE) of manufactured parts. The lack of simple, reliable, and inexpensive methods for NDE of AM parts is a significant obstacle to wider adoption of AM parts. Electromechanical impedance measurements have been investigated as a means to detect manufacturing defects in AM parts. Impedance-based NDE utilizes piezoelectric wafers as collocated sensors and actuators. Taking advantage of the coupled electromechanical characteristics of piezoelectric materials, the mechanical characteristics of the part under test can be inferred from the electrical impedance of the piezoelectric wafer. Previous efforts have used piezoelectric wafers bonded directly to the part under test, which imposes several challenges regarding the applicability and robustness of the technique. This paper investigates the use of an instrumented clamp as a solution for measuring the electromechanical impedance of the part under test. The effectiveness of this approach in detecting manufacturing defects is compared to directly bonded wafers.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1115/1.4065276
