{
    "id": 921,
    "date": "2023-04-26T22:51:23",
    "date_gmt": "2023-04-26T15:51:23",
    "guid": {
        "rendered": "https:\/\/technica.ai\/?p=921"
    },
    "modified": "2025-08-01T15:34:31",
    "modified_gmt": "2025-08-01T08:34:31",
    "slug": "wafer-defect-classification",
    "status": "publish",
    "type": "post",
    "link": "https:\/\/technica.ai\/en\/case-studies\/wafer-defect-classification\/",
    "title": {
        "rendered": "Wafer Defect Classification"
    },
    "content": {
        "rendered": "<h4>Challenge<\/h4>\n<p>It is required to automatically identify the cause of defects using scanning electron microscope (SEM) images showing defects detected by semiconductor wafer defect inspection equipment.<\/p>\n<p>Although machine learning is an effective method, it has the following problems.<\/p>\n<ul>\n<li>Imbalanced data<\/li>\n<li>Lack of training data<\/li>\n<li>Differences of images in a class may outweigh differences of images between classes<\/li>\n<\/ul>\n<h4>Solution<\/h4>\n<p>Our solution is a combination of many methods: Anomaly Detection, Object Classification, Region Analysis<\/p>\n<ul>\n<li>The Anomaly Detection algorithm only needs good samples to train, so the lack of training data does not affect the result.<\/li>\n<li>Next, we based on the characteristics of the error area to distinguish two main types of defects, scratches or foreign bodies.<\/li>\n<li>With scratches, the severity of the defect is assessed by size. With foreign bodies, use the Object Classification algorithm to distinguish the type of foreign body.<\/li>\n<\/ul>\n<h4>Outcome<\/h4>\n<p>Automatically classifies known defect causes and detects unknown defects that do not belong to any class. Achieved 96% classification accuracy.<\/p>",
        "protected": false
    },
    "excerpt": {
        "rendered": "<p>Automatically classifies known defect causes and detects unknown defects that do not belong to any class. Achieved 96% classification accuracy.<\/p>",
        "protected": false
    },
    "author": 2,
    "featured_media": 342,
    "comment_status": "closed",
    "ping_status": "open",
    "sticky": false,
    "template": "",
    "format": "standard",
    "meta": {
        "inline_featured_image": false,
        "footnotes": ""
    },
    "categories": [
        9
    ],
    "tags": [],
    "class_list": [
        "post-921",
        "post",
        "type-post",
        "status-publish",
        "format-standard",
        "has-post-thumbnail",
        "hentry",
        "category-case-studies"
    ],
    "acf": [],
    "_links": {
        "self": [
            {
                "href": "https:\/\/technica.ai\/en\/wp-json\/wp\/v2\/posts\/921",
                "targetHints": {
                    "allow": [
                        "GET"
                    ]
                }
            }
        ],
        "collection": [
            {
                "href": "https:\/\/technica.ai\/en\/wp-json\/wp\/v2\/posts"
            }
        ],
        "about": [
            {
                "href": "https:\/\/technica.ai\/en\/wp-json\/wp\/v2\/types\/post"
            }
        ],
        "author": [
            {
                "embeddable": true,
                "href": "https:\/\/technica.ai\/en\/wp-json\/wp\/v2\/users\/2"
            }
        ],
        "replies": [
            {
                "embeddable": true,
                "href": "https:\/\/technica.ai\/en\/wp-json\/wp\/v2\/comments?post=921"
            }
        ],
        "version-history": [
            {
                "count": 1,
                "href": "https:\/\/technica.ai\/en\/wp-json\/wp\/v2\/posts\/921\/revisions"
            }
        ],
        "predecessor-version": [
            {
                "id": 5711,
                "href": "https:\/\/technica.ai\/en\/wp-json\/wp\/v2\/posts\/921\/revisions\/5711"
            }
        ],
        "wp:featuredmedia": [
            {
                "embeddable": true,
                "href": "https:\/\/technica.ai\/en\/wp-json\/wp\/v2\/media\/342"
            }
        ],
        "wp:attachment": [
            {
                "href": "https:\/\/technica.ai\/en\/wp-json\/wp\/v2\/media?parent=921"
            }
        ],
        "wp:term": [
            {
                "taxonomy": "category",
                "embeddable": true,
                "href": "https:\/\/technica.ai\/en\/wp-json\/wp\/v2\/categories?post=921"
            },
            {
                "taxonomy": "post_tag",
                "embeddable": true,
                "href": "https:\/\/technica.ai\/en\/wp-json\/wp\/v2\/tags?post=921"
            }
        ],
        "curies": [
            {
                "name": "wp",
                "href": "https:\/\/api.w.org\/{rel}",
                "templated": true
            }
        ]
    }
}