Fully automated SEM image analysis

Edwards, R. M.; Lebiedzik, J.; Stone, Garvin

doi:10.1002/sca.4950080505

“…The element like width and location are retrieved from the vein features and combined using concatenation. In the process of recognition, features are produced from the test pictures and matched with the elements being encrypted in databases based on the application of the Euclidean distance measurement [10].…”

Section: Adaptive Histogram Equalization (Ahe)mentioning

confidence: 99%

Pattern Recognition for Veins and Feature Extraction Techniques

Welland

¹

2021

International Journal of Advanced Information and Communication Technology

0

View full text Add to dashboard Cite

Among the biometric system approaches, the Vein Pattern Recognition (VPR) is one of the rapidly expanding aspects of medical image enhancement and scanning. Whereas the ideology behind the technique is modest, there are some problems reported throughout the implementation and designing of vein scanning devices, regarding the hardware lighting framework and the real algorithms utilized for the implementation and processing of a number of images. To maintain the scanning issues to minimize measure, the images from the cameras have to be nearly noiseless and the algorithms have to be capable of detecting the patterns of the veins in different actual-life scenarios. Most implementations of the approach have been commercialized which makes it necessary to launch a system which can extract, analyse and detect the precise human vein patterns whereas maintaining the minimal costs and minimizing the computation requirements of image processing algorithms. Based on that this provides a critical survey of pattern recognition for VPR. Moreover, it provides the hardware processing and implementation remedies based on the approach.

show abstract

“…It might be considered that the DL categorizer is the development of the common Artificial Neural Network (ANN) since it included more layers compared to ANN. Moreover, it is known as a form of representational ML since every layer changes the input information from the initial layer into a novel representation as a significantly higher level of abstraction [6]. This permits the framework to master both international and localized connections of the complete data in a more hierarchical framework.…”

Section: Deep Learning Categorizermentioning

confidence: 99%

Big Data Techniques for Clinical Image Analysis

Ying

¹

,

Yu

²

2021

International Journal of Advanced Information and Communication Technology

0

View full text Add to dashboard Cite

The process of extracting clinical images, for instance, heart images from cameras is full of noises and complexities. As such, the general expenditure for processing these images like resources and time is significantly high, mostly for complex and large amounts of data. In that case, this research contribution utilizes the machine vision-centred method to effectively address these issues. The method significantly incorporates four essential stages with various forms of algorithms to handle the clinical heart images. In the first stage, the smoothing algorithm is utilized to minimize some form of noise. In the second stage, the filtering algorithms are applied for the analysis of images to effectively identify the targeted region. In the third stage, more developed algorithm is utilized to evaluate the image characteristics in the targeted region, identifying the basic image outline. Lastly, the reduction algorithm is meant to transform the original image into significantly precise and smooth pictures. The experimental findings indicate that the machine vision-centred clinical image analysis algorithm might significantly extract fundamental data and attain the most reliable results, contrasting with the most ancient image analysis techniques.

show abstract

“…In that case, the image contrast, C x y, is identified by the texture data f u x y. (6) Whereby is considered as (7) Based on the application of the image development algorithms, the comparison development of the heart image reflects not just the international pixel data, but also evaluates the localized details of the image data. The detailed partial data is reflected based on the weighted function which is considered to be more enhanced by the pixel of images and their intensity.…”

Section: Step 3: Image Enhancementmentioning

confidence: 99%

Big Data Techniques for Clinical Image Analysis

Ying

¹

,

Yu

²

2020

JMIC

0

View full text Add to dashboard Cite

The process of extracting clinical images, for instance, heart images from cameras is full of noises and complexities. As such, the general expenditure for processing these images like resources and time is significantly high, mostly for complex and large amounts of data. In that case, this research contribution utilizes the machine vision-centred method to effectively address these issues. The method significantly incorporates four essential stages with various forms of algorithms to handle the clinical heart images. In the first stage, the smoothing algorithm is utilized to minimize some form of noise. In the second stage, the filtering algorithms are applied for the analysis of images to effectively identify the targeted region. In the third stage, more developed algorithm is utilized to evaluate the image characteristics in the targeted region, identifying the basic image outline. Lastly, the reduction algorithm is meant to transform the original image into significantly precise and smooth pictures. The experimental findings indicate that the machine vision-centred clinical image analysis algorithm might significantly extract fundamental data and attain the most reliable results, contrasting with the most ancient image analysis techniques.

show abstract

Fully automated SEM image analysis

Cited by 13 publications

References 2 publications

Pattern Recognition for Veins and Feature Extraction Techniques

Pattern Recognition for Veins and Feature Extraction Techniques

Big Data Techniques for Clinical Image Analysis

Big Data Techniques for Clinical Image Analysis

Contact Info

Product

Resources

About