computer science specially image processing and image segmentation
Hi, the basic idea of Ostu is follows
Initially it works directly on the gray level histogram
Find the threshold that minimizes the weighted within-class variance
for detailed information check the following link
https://www.google.co.in/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&uact=8&ved=0ahUKEwia-f6OnpjMAhUFv5QKHV2TDUMQFgggMAA&url=https%3A%2F%2Fwebdocs.cs.ualberta.ca%2F~nray1%2FCMPUT605%2Ftrack3_papers%2Fotsu.ppt&usg=AFQjCNHoBt2EXNB3sx2uvqkRFrJe-fwzaA&bvm=bv.119745492,d.dGo
Hi, the basic idea of Ostu is follows
Initially it works directly on the gray level histogram
Find the threshold that minimizes the weighted within-class variance
for detailed information check the following link
https://www.google.co.in/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&uact=8&ved=0ahUKEwia-f6OnpjMAhUFv5QKHV2TDUMQFgggMAA&url=https%3A%2F%2Fwebdocs.cs.ualberta.ca%2F~nray1%2FCMPUT605%2Ftrack3_papers%2Fotsu.ppt&usg=AFQjCNHoBt2EXNB3sx2uvqkRFrJe-fwzaA&bvm=bv.119745492,d.dGo
Simply, it is an image thresholding algorithm for reduction of a gray level image to a binary image.
Segmentation is often considered to be the first step in image analysis.In general, autonomous segmentation is one of the most difficult tasks in digital image processing. Otsu’s thresholding chooses the threshold to minimize the intraclass variance of the thresholded black and white pixels.
U can take a look at the links below which are some tutorials providing the basics of OTSU thresholding:
http://www.math.tau.ac.il/~turkel/notes/otsu.pdf
http://web-ext.u-aizu.ac.jp/course/bmclass/documents/otsu1979.pdf
https://en.wikipedia.org/wiki/Otsu%27s_method
This is a good question.
In addition to very good answers already given, there is a bit more to consider. Articles on Otsu's method are available here on RG. A good introduction to Otsu's method is given in
https://www.researchgate.net/publication/221196493_Image_Segmentation_Based_on_2D_Otsu_Method_with_Histogram_Analysis
See also:
https://www.researchgate.net/publication/238734218_An_improved_image_segmentation_algorithm_based_on_Otsu_method
https://www.researchgate.net/publication/224592999_Otsu_Method_and_K-means
https://www.researchgate.net/publication/259849226_11_IJFTET_-_Vol_4-Issue_1_Improved_Otsu's_Thresholding_Techniques_for_HSI_Color_Image_Segmentation
https://www.researchgate.net/publication/224085716_A_Fast_2D_Otsu_Thresholding_Algorithm_Based_on_Improved_Histogram
Conference Paper Image Segmentation Based on 2D Otsu Method with Histogram Analysis
Article An improved image segmentation algorithm based on Otsu method
Conference Paper Otsu Method and K-means
Article 11. IJFTET - Vol. 4-Issue 1_Improved Otsu's Thresholding Tec...
Conference Paper A Fast 2D Otsu Thresholding Algorithm Based on Improved Histogram
Response by Ranjit is straight-forward but abstract. Otsu's algorithm is more sophisticated way of image thesholding as compared to traditional mean-based split of gray-level image into binary image. This sophistication comes from minimizing the weighted within-class variance and maximizing between-class variance, as quoted by Chethan and Singha.
A simple and fitting example can be found at:
http://www.labbookpages.co.uk/software/imgProc/otsuThreshold.html
Than you much my colleague Abdul Kayom Md Khairuzzaman for your interesting answer.
Even though, Otsu method is quite fast. It has one drawback, it is required uniform illumination which is difficult in practice. You can use it in addition to any other method.
Can someone help me please, I would apply the otsu's method to calculate the adaptive threshold which can be used in threshold based fall detection
methods to improve the accuracy of fall detection. Is it possible ??
this useful
https://pdfs.semanticscholar.org/ee8b/aed71d46b4d8747c8f2c73ba73f7c4a06130.pdf
http://engr.case.edu/merat_francis/eecs490f07/lectures/lecture22.pdf
Otsu's thresholding method involves iterating through all the possible threshold values and calculating a measure of spread for the pixel levels each side of the threshold, i.e. the pixels that either fall in foreground or background.
http://www.labbookpages.co.uk/software/imgProc/otsuThreshold.html
- Badges
- Science topic
- Similar topics
- Engineering
- Electrical Engineering
- Signal Processing
- Image Processing