18045598. IMAGE ANOMALY DETECTION BY ENHANCING PATCHED FEATURES simplified abstract (International Business Machines Corporation)
Contents
- 1 IMAGE ANOMALY DETECTION BY ENHANCING PATCHED FEATURES
- 1.1 Organization Name
- 1.2 Inventor(s)
- 1.3 IMAGE ANOMALY DETECTION BY ENHANCING PATCHED FEATURES - A simplified explanation of the abstract
- 1.4 Simplified Explanation
- 1.5 Potential Applications
- 1.6 Problems Solved
- 1.7 Benefits
- 1.8 Potential Commercial Applications
- 1.9 Possible Prior Art
- 1.10 Original Abstract Submitted
IMAGE ANOMALY DETECTION BY ENHANCING PATCHED FEATURES
Organization Name
International Business Machines Corporation
Inventor(s)
TADANOBU Inoue of Yokohama (JP)
Takayuki Katsuki of Tokyo (JP)
IMAGE ANOMALY DETECTION BY ENHANCING PATCHED FEATURES - A simplified explanation of the abstract
This abstract first appeared for US patent application 18045598 titled 'IMAGE ANOMALY DETECTION BY ENHANCING PATCHED FEATURES
Simplified Explanation
The patent application relates to accurate anomaly detection in images using patched features. Here is a simplified explanation of the abstract:
- An extraction component extracts features from patches of an image using a pretrained convolutional neural network.
- A feature mapping component combines the features from multiple layers to create a one-dimensional feature vector for each patch.
- A cropping component performs center cropping on the feature map.
- A calculation component determines the distance to a feature distribution mean for each patch.
Potential Applications
This technology could be applied in various fields such as medical imaging for detecting anomalies in X-rays or MRIs, security systems for identifying suspicious objects in surveillance footage, and quality control in manufacturing for inspecting products for defects.
Problems Solved
This technology solves the problem of accurately detecting anomalies in images by utilizing patched features and a pretrained convolutional neural network. It improves the efficiency and accuracy of anomaly detection compared to traditional methods.
Benefits
The benefits of this technology include improved accuracy in anomaly detection, faster processing of images, and the ability to adapt to different types of anomalies in various applications.
Potential Commercial Applications
Potential commercial applications of this technology include developing software for medical imaging companies, security firms, and manufacturing companies to integrate accurate anomaly detection capabilities into their systems.
Possible Prior Art
One possible prior art for this technology could be the use of convolutional neural networks for image feature extraction and anomaly detection. Researchers and companies have been exploring similar techniques in the field of computer vision and image processing.
Unanswered Questions
How does this technology compare to existing anomaly detection methods in terms of accuracy and efficiency?
This article does not provide a direct comparison with existing anomaly detection methods to evaluate its performance in terms of accuracy and efficiency.
What are the limitations or challenges of implementing this technology in real-world applications?
The article does not address the potential limitations or challenges of implementing this technology in practical, real-world scenarios.
Original Abstract Submitted
One or more systems, devices, computer program products, and/or computer-implemented methods provided herein relate to accurate anomaly detection in images using patched features. According to an embodiment, an extraction component can extract multiple layers of features from one or more patches of an image using a pretrained convolutional neural network (CNN). A feature mapping component can concatenate the features from the multiple layers to generate a tensor feature map comprising a one-dimensional feature vector for respective patches. A cropping component can perform center cropping on the tensor feature map. A calculation component can calculate a distance to a feature distribution mean for respective patches.