Other Scholarly Content
 

A Novel method to detecting lines on an image with high noise density

Public Deposited

Downloadable Content

Download PDF
https://ir.library.oregonstate.edu/concern/defaults/5999n494b

Descriptions

Attribute NameValues
Creator
Abstract
  • We developed an integration-based line detection algorithm. Existing line detection methods such as the Hough Transformation (HT) and its variants are insensitive to image noise. The reason is that HT finds lines by calculating the gradient of the image and assumes that the region where the gradient is the steepest is where lines exist. This is problematic because if an image has an extremely noisy region, then HT can produce false positive results. Using filters to remove noise on images can increase computational complexity. There are existing line detection algorithms based on template matching that are robust against noise. However, those algorithms are complex and hard to implement. Our method identifies lines by calculating the correlation score between a set of template line images and the raw image. We calculate the correlation score by multiplying each pixel between the template line image and the raw image and summing their product, or integrating each pixel on the raw image. Our algorithm is simple to implement and requires no application of noise filters onto a noisy image. Additionally, our algorithm removes the necessity for users to use segmentation techniques such as Canny edge detection. We were able to use our algorithm to extract collagen fibers from a noisy image produced by a confocal microscope.
Contributor
Resource Type
Date Available
Date Issued
Degree Level
Degree Name
Degree Field
Advisor
Academic Affiliation
Non-Academic Affiliation
Subject
Rights Statement
Related Items
  • Poster
Publisher
Peer Reviewed
Language
Replaces

Relationships

Parents:

This work has no parents.

In Collection:

Items

Thumbnail Title Date Uploaded Visibility Actions
file details Dan_Lin_physics_seminar_presentation.pdf 2017-07-21 Public Download