A Survey on Segmentation Techniques for Image Processing

Jasim, Wala'a Nsaif; Mohammed, Reham H.

doi:10.37917/ijeee.17.2.10

Cited by 19 publications

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thresholding techniques do not penalize false positive, or false negative pixels which makes achieving a stable threshold extremely difficult [45]. Consequently, these methods, require fine-tuning of threshold values to account for image quality differences [46], which limits their generalization to real clinical settings.…”

Section: Related Workmentioning

confidence: 99%

Full-Resolution Lung Nodule Localization From Chest X-Ray Images Using Residual Encoder-Decoder Networks

Horry,

Chakraborty,

Pradhan

et al. 2023

IEEE Access

View full text Add to dashboard Cite

Lung cancer is the leading cause of cancer death, and early diagnosis is associated with a positive prognosis. Chest X-ray (CXR) provides an inexpensive imaging mode for lung cancer diagnosis. Computer vision algorithms have previously been proposed to assist human radiologists in this task; however, leading studies use down-sampled images and computationally expensive methods with unproven generalization. In contrast, this study localizes lung nodules from CXR images using efficient encoder-decoder neural networks that have been crafted to process full resolution input images, thereby avoiding signal loss resulting from down-sampling. Encoder-decoder networks are trained and tested using the Japanese Society of Radiological Technology dataset. The networks are used to localize lung nodules from an independent CXR dataset. These experiments allow for the determination of the optimal network depth, image resolution, and pre-processing pipeline for generalized lung nodule localization. We find that more subtle nodules are detected in earlier training epochs. Therefore, we propose a novel self-ensemble model from three consecutive epochs centered on the validation optimum. This ensemble achieved a sensitivity of 85% in 10-fold internal testing with false positives of 8 per image. A sensitivity of 81% is achieved at a false positive rate of 6 following morphological false positive reduction. This result is comparable to more computationally complex systems, but with a sub-second inference time that is faster than other methods presented in the literature. The proposed algorithm achieved excellent generalization results against a challenging external dataset with a sensitivity of 77% at a false positive rate of 7.6.

show abstract

Section: Related Workmentioning

confidence: 99%

Full-Resolution Lung Nodule Localization From Chest X-Ray Images Using Residual Encoder-Decoder Networks

Horry,

Chakraborty,

Pradhan

et al. 2023

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Depending on the similarity the growth starts with a pixel and continues to add pixels to the area, repeating until all the pixels belong to a certain area (Zaitoun & Aqel, 2015). This methodology is considered Statistical Region Merging (SRM), Seeded Region Growing (SRG), and Unseeded Region Growing (Jasim & Mohammed, 2021).…”

Section: Region Growingmentioning

confidence: 99%

“…There are some steps to perform the Unseeded Region Growing technique, and the principal steps can be brief (Jasim & Mohammed, 2021):…”

Section: Unseeded Region Growing Algorithmmentioning

confidence: 99%

Review on Region-Based Segmentation Using Watershed and Region Growing Techniques and their Applications in Different Fields

khalid

2022

JLMA

View full text Add to dashboard Cite

In digital image processing and computer vision, segmentation operation for an image refers to dividing an image into multiple image segments, and the significant purpose of segmentation operation is to depict an image in a way so that the analysis process of the objects of interest is easier and more accurate. The region-based segmentation scheme act for finding similarities between adjacent pixels to detect each region that constructs the image. Similarity scales have based on different features, in a grayscale image, the scale may be referred to as textures and other spatial appearances, and also the variance in intensity of a region and so on. Significantly, many applications in different fields involved region-based segmentation for instance remote sensing, medical application, and others for recognizing interesting objects in an image. In this paper, two techniques for segmentation operation in region-based which are region growing and watershed are reviewed.

show abstract

Computer vision‐based automated defect detection in ceramic bricks

Kataev,

Bulysheva

2024

Syst Res Behav Sci

View full text Add to dashboard Cite

Nowadays, the development of cost‐effective, data‐driven technological processes using telecommunication technologies is essential. One of the focuses is on automating the process of evaluating the manufactured goods' quality. Vision‐based technology is now becoming increasingly used for monitoring purposes. Despite its advancements, computer vision technology has practical limitations. These include the physical characteristics of the measuring process, features specific to the technological procedures, and constraints related to software and mathematical algorithms. Among the cutting‐edge approaches, optical methods combined with neural network algorithms (NN) stand out. This significance is particularly evident because numerous industries continue to depend on manual defect identification methods, which are labour intensive, slow, and subject to human subjectivity. The article introduces a novel approach based on computer vision methods. It outlines an automated optical inspection system designed to detect defects in bricks on a transport belt during the production process. The article presents the processing algorithms used and discusses the results obtained.

show abstract

A Survey on Segmentation Techniques for Image Processing

Cited by 19 publications

References 0 publications

Full-Resolution Lung Nodule Localization From Chest X-Ray Images Using Residual Encoder-Decoder Networks

Full-Resolution Lung Nodule Localization From Chest X-Ray Images Using Residual Encoder-Decoder Networks

Review on Region-Based Segmentation Using Watershed and Region Growing Techniques and their Applications in Different Fields

Computer vision‐based automated defect detection in ceramic bricks

Contact Info

Product

Resources

About