Feature Extraction From Analog Wafermaps: A Comparison of Classical Image Processing and a Deep Generative Model

Tiago Filipe Teixeira dos Santos, Stefan Schrunner, Bernhard Geiger, Olivia Pfeiler, Anja Zernig, Andre Kaestner, Roman Kern

Publikation: Beitrag in einer FachzeitschriftArtikelForschungBegutachtung

Abstract

Semiconductor manufacturing is a highly innovative branch of industry, where a high degree of automation has already been achieved. For example, devices tested to be outside of their specifications in electrical wafer test are automatically scrapped. In this paper, we go one step further and analyze test data of devices still within the limits of the specification, by exploiting the information contained in the analog wafermaps. To that end, we propose two feature extraction approaches with the aim to detect patterns in the wafer test dataset. Such patterns might indicate the onset of critical deviations in the production process. The studied approaches are: 1) classical image processing and restoration techniques in combination with sophisticated feature engineering and 2) a data-driven deep generative model. The two approaches are evaluated on both a synthetic and a real-world dataset. The synthetic dataset has been modeled based on real-world patterns and characteristics. We found both approaches to provide similar overall evaluation metrics. Our in-depth analysis helps to choose one approach over the other depending on data availability as a major aspect, as well as on available computing power and required interpretability of the results.

Originalspracheenglisch
Aufsatznummer8689102
Seiten (von - bis)190-198
Seitenumfang9
FachzeitschriftIEEE transactions on semiconductor manufacturing
Jahrgang32
Ausgabenummer2
DOIs
PublikationsstatusVeröffentlicht - Mai 2019

Fingerprint

pattern recognition
image processing
Feature extraction
Image processing
analogs
Specifications
specifications
Image reconstruction
wafers
Automation
Availability
Semiconductor materials
automation
restoration
availability
manufacturing
industries
engineering
deviation
Industry

Schlagwörter

    ASJC Scopus subject areas

    • !!Electronic, Optical and Magnetic Materials
    • !!Condensed Matter Physics
    • !!Electrical and Electronic Engineering
    • !!Industrial and Manufacturing Engineering

    Dies zitieren

    Feature Extraction From Analog Wafermaps: A Comparison of Classical Image Processing and a Deep Generative Model. / Teixeira dos Santos, Tiago Filipe; Schrunner, Stefan; Geiger, Bernhard; Pfeiler, Olivia; Zernig, Anja; Kaestner, Andre; Kern, Roman.

    in: IEEE transactions on semiconductor manufacturing, Jahrgang 32, Nr. 2, 8689102, 05.2019, S. 190-198.

    Publikation: Beitrag in einer FachzeitschriftArtikelForschungBegutachtung

    @article{2a36d4ce1bb442c8922604e4c9ea3f44,
    title = "Feature Extraction From Analog Wafermaps: A Comparison of Classical Image Processing and a Deep Generative Model",
    abstract = "Semiconductor manufacturing is a highly innovative branch of industry, where a high degree of automation has already been achieved. For example, devices tested to be outside of their specifications in electrical wafer test are automatically scrapped. In this paper, we go one step further and analyze test data of devices still within the limits of the specification, by exploiting the information contained in the analog wafermaps. To that end, we propose two feature extraction approaches with the aim to detect patterns in the wafer test dataset. Such patterns might indicate the onset of critical deviations in the production process. The studied approaches are: 1) classical image processing and restoration techniques in combination with sophisticated feature engineering and 2) a data-driven deep generative model. The two approaches are evaluated on both a synthetic and a real-world dataset. The synthetic dataset has been modeled based on real-world patterns and characteristics. We found both approaches to provide similar overall evaluation metrics. Our in-depth analysis helps to choose one approach over the other depending on data availability as a major aspect, as well as on available computing power and required interpretability of the results.",
    keywords = "Automation, data processing, feature extraction, unsupervised learning",
    author = "{Teixeira dos Santos}, {Tiago Filipe} and Stefan Schrunner and Bernhard Geiger and Olivia Pfeiler and Anja Zernig and Andre Kaestner and Roman Kern",
    year = "2019",
    month = "5",
    doi = "10.1109/TSM.2019.2911061",
    language = "English",
    volume = "32",
    pages = "190--198",
    journal = "IEEE transactions on semiconductor manufacturing",
    issn = "0894-6507",
    publisher = "Institute of Electrical and Electronics Engineers",
    number = "2",

    }

    TY - JOUR

    T1 - Feature Extraction From Analog Wafermaps: A Comparison of Classical Image Processing and a Deep Generative Model

    AU - Teixeira dos Santos, Tiago Filipe

    AU - Schrunner, Stefan

    AU - Geiger, Bernhard

    AU - Pfeiler, Olivia

    AU - Zernig, Anja

    AU - Kaestner, Andre

    AU - Kern, Roman

    PY - 2019/5

    Y1 - 2019/5

    N2 - Semiconductor manufacturing is a highly innovative branch of industry, where a high degree of automation has already been achieved. For example, devices tested to be outside of their specifications in electrical wafer test are automatically scrapped. In this paper, we go one step further and analyze test data of devices still within the limits of the specification, by exploiting the information contained in the analog wafermaps. To that end, we propose two feature extraction approaches with the aim to detect patterns in the wafer test dataset. Such patterns might indicate the onset of critical deviations in the production process. The studied approaches are: 1) classical image processing and restoration techniques in combination with sophisticated feature engineering and 2) a data-driven deep generative model. The two approaches are evaluated on both a synthetic and a real-world dataset. The synthetic dataset has been modeled based on real-world patterns and characteristics. We found both approaches to provide similar overall evaluation metrics. Our in-depth analysis helps to choose one approach over the other depending on data availability as a major aspect, as well as on available computing power and required interpretability of the results.

    AB - Semiconductor manufacturing is a highly innovative branch of industry, where a high degree of automation has already been achieved. For example, devices tested to be outside of their specifications in electrical wafer test are automatically scrapped. In this paper, we go one step further and analyze test data of devices still within the limits of the specification, by exploiting the information contained in the analog wafermaps. To that end, we propose two feature extraction approaches with the aim to detect patterns in the wafer test dataset. Such patterns might indicate the onset of critical deviations in the production process. The studied approaches are: 1) classical image processing and restoration techniques in combination with sophisticated feature engineering and 2) a data-driven deep generative model. The two approaches are evaluated on both a synthetic and a real-world dataset. The synthetic dataset has been modeled based on real-world patterns and characteristics. We found both approaches to provide similar overall evaluation metrics. Our in-depth analysis helps to choose one approach over the other depending on data availability as a major aspect, as well as on available computing power and required interpretability of the results.

    KW - Automation

    KW - data processing

    KW - feature extraction

    KW - unsupervised learning

    UR - http://www.scopus.com/inward/record.url?scp=85065391683&partnerID=8YFLogxK

    U2 - 10.1109/TSM.2019.2911061

    DO - 10.1109/TSM.2019.2911061

    M3 - Article

    VL - 32

    SP - 190

    EP - 198

    JO - IEEE transactions on semiconductor manufacturing

    JF - IEEE transactions on semiconductor manufacturing

    SN - 0894-6507

    IS - 2

    M1 - 8689102

    ER -