View source for 20230169762. METHOD FOR OPTIMIZING DETECTION OF ABNORMALITIES IN IMAGES, TERMINAL DEVICE, AND COMPUTER READABLE STORAGE MEDIUM APPLYING THE METHOD simplified abstract (HON HAI PRECISION INDUSTRY CO., LTD.)

=METHOD FOR OPTIMIZING DETECTION OF ABNORMALITIES IN IMAGES, TERMINAL DEVICE, AND COMPUTER READABLE STORAGE MEDIUM APPLYING THE METHOD=


==Organization Name==

[[:Category:HON HAI PRECISION INDUSTRY CO., LTD.|HON HAI PRECISION INDUSTRY CO., LTD.]]


[[Category:HON HAI PRECISION INDUSTRY CO., LTD.]]

==Inventor(s)==

[[:Category:CHUNG-YU Wu of New Taipei (TW)|CHUNG-YU Wu of New Taipei (TW)]][[Category:CHUNG-YU Wu of New Taipei (TW)]]

[[:Category:GUO-CHIN Sun of New Taipei (TW)|GUO-CHIN Sun of New Taipei (TW)]][[Category:GUO-CHIN Sun of New Taipei (TW)]]

[[:Category:CHIH-TE Lu of New Taipei (TW)|CHIH-TE Lu of New Taipei (TW)]][[Category:CHIH-TE Lu of New Taipei (TW)]]

==METHOD FOR OPTIMIZING DETECTION OF ABNORMALITIES IN IMAGES, TERMINAL DEVICE, AND COMPUTER READABLE STORAGE MEDIUM APPLYING THE METHOD - A simplified explanation of the abstract==


This abstract first appeared for US patent application 20230169762 titled 'METHOD FOR OPTIMIZING DETECTION OF ABNORMALITIES IN IMAGES, TERMINAL DEVICE, AND COMPUTER READABLE STORAGE MEDIUM APPLYING THE METHOD

==Simplified Explanation==

The patent application describes a method for optimizing the detection of abnormalities in product images using a generative adversarial network (GAN). Here are the key points:

* The method involves generating two types of images: a first image similar to training images and a second image similar to testing images.
* Both normal images and images showing abnormalities are inputted into a GAN.
* The GAN determines a similarity ratio between the first image and the training image, and generates a parameter based on this ratio to adjust the GAN.
* A similarity ratio is also determined between the second image and the testing image.
* If the second similarity ratio is larger than a specified threshold value, the testing image is deemed normal. If it is less than or equal to the threshold, the testing image is deemed to reveal abnormalities.
* The method can be implemented on a terminal device and stored on a computer-readable storage medium.

Potential applications of this technology:

* Quality control in manufacturing: The method can be used to detect abnormalities in product images, helping to identify faulty or defective products during the manufacturing process.
* Medical imaging: The method can be applied to medical images to detect abnormalities or anomalies, aiding in the diagnosis and treatment of diseases.
* Security and surveillance: The method can be used to analyze images captured by security cameras or surveillance systems, identifying any abnormal or suspicious activities.

Problems solved by this technology:

* Efficient detection of abnormalities: The method provides a way to automatically and accurately detect abnormalities in images, reducing the need for manual inspection and improving efficiency.
* Early detection of issues: By analyzing images in real-time, the method enables the early detection of abnormalities or anomalies, allowing for prompt action to be taken.
* Standardization of detection: The use of a GAN helps to standardize the detection process, ensuring consistent and reliable results across different images and scenarios.

Benefits of this technology:

* Improved accuracy: By leveraging the power of a GAN, the method can achieve high accuracy in detecting abnormalities, minimizing false positives and false negatives.
* Time and cost savings: Automating the detection process reduces the need for manual inspection, saving time and labor costs.
* Scalability: The method can be applied to a wide range of industries and use cases, making it a versatile solution for detecting abnormalities in various types of images.


==Original Abstract Submitted==

a method of optimizing the detection of abnormalities in images of products generates a first image similar to training images and a second image similar to testing images with normal images and the images showing abnormalities inputted into a generative adversarial network (gan). the gan determines a first similarity ratio between the first image and the training image and generates a parameter based on the first similarity ratio for adjusting the gan. a second similarity ratio between the second image and the testing image is determined. the testing image is deemed a normal image when the second similarity ratio is larger than the specified threshold value, and deemed to be an image revealing abnormalities when the second similarity ratio is less than or equal to the specified threshold value. a terminal device and a computer readable storage medium applying the method are also provided.



[[Category:G06V10/98]]
[[Category:G06V10/74]]
[[Category:G06V10/77]]
[[Category:G06V10/82]]
[[Category:G06N3/04]]