18070744. MODELING FOR INDEXING AND SEMICONDUCTOR DEFECT IMAGE RETRIEVAL simplified abstract (Applied Materials, Inc.)

MODELING FOR INDEXING AND SEMICONDUCTOR DEFECT IMAGE RETRIEVAL

Organization Name

Applied Materials, Inc.

Inventor(s)

Yuanhong Guo of Mountain View CA (US)

Sachin Dangayach of San Jose CA (US)

Rahul Reddy Komatireddi of Hyderabad (IN)

Tianyuan Wu of San Jose CA (US)

MODELING FOR INDEXING AND SEMICONDUCTOR DEFECT IMAGE RETRIEVAL - A simplified explanation of the abstract

This abstract first appeared for US patent application 18070744 titled 'MODELING FOR INDEXING AND SEMICONDUCTOR DEFECT IMAGE RETRIEVAL

Simplified Explanation

The subject matter of this specification involves a method for processing image frames to detect substrate processing defects. The method includes storing feature vectors of previously processed image frames, receiving new image data, determining feature vectors for the new data, selecting similar feature vectors from the stored ones, and performing actions based on the selected feature vectors.

• Method for processing image frames to detect substrate processing defects:

   - Store feature vectors of previously processed image frames
   - Receive new image data
   - Determine feature vectors for the new data
   - Select similar feature vectors from the stored ones
   - Perform actions based on the selected feature vectors

Potential Applications

This technology can be applied in industries where substrate processing defects need to be detected and corrected, such as semiconductor manufacturing, pharmaceutical production, and quality control in various manufacturing processes.

Problems Solved

This technology helps in automating the detection of substrate processing defects, reducing manual inspection time, and improving the overall quality control process in manufacturing industries.

Benefits

The benefits of this technology include increased efficiency in defect detection, improved accuracy in identifying substrate processing defects, and overall cost savings due to reduced manual labor.

Potential Commercial Applications

The potential commercial applications of this technology include integration into automated inspection systems for manufacturing processes, quality control software for various industries, and defect detection tools for substrate processing facilities.

Possible Prior Art

One possible prior art for this technology could be existing image processing algorithms used in quality control systems for manufacturing processes. These algorithms may have similarities in detecting defects but may not specifically focus on substrate processing defects.

How does this technology compare to manual defect detection methods in terms of accuracy and efficiency?

This article does not provide a direct comparison between this technology and manual defect detection methods. Manual methods may have limitations in terms of accuracy and efficiency compared to automated systems like the one described in the patent application.

What are the potential limitations or challenges in implementing this technology in real-world manufacturing environments?

The article does not address the potential limitations or challenges in implementing this technology in real-world manufacturing environments. Some challenges could include integration with existing systems, scalability for large-scale manufacturing operations, and training requirements for users.

Original Abstract Submitted

The subject matter of this specification can be implemented in, among other things, methods, systems, computer-readable storage medium. A method can include a processing device storing a plurality of feature vectors representative of previously processed image frames that correspond to various substrate processing defects. The method further includes receiving first image data comprising one or more image frames indicative of a first substrate processing defect. The method further includes determining a first feature vector corresponding to the first image data. The method further includes determining a selection of the plurality of feature vectors based on a proximity between the first feature vector and each of the selection of the plurality of feature vectors. The method further includes determining second image data comprising one or more image frames corresponding to the selection of the plurality of embedding vectors and performing an action based on determining the second image data.