OD position

[hahabingo]

hello, Is the location of optic disc in this article based on the selection of the highest point of pixel brightness? I find it didn't work to some images.

[babloogpb1]

Hii, The centre of od is not the brightest pixel location. If you refer to the paper, the authors actually mentioned a procedure to find the centroid of od using some distance function and a close hole operator. We didn't implement that part in the code, but we computed the centroid by considering the brightest region(od) and finding its centre.

[hahabingo]

Thank you very much for your reply. I got it. Have you implemented his method in your code now? 

------------------ 原始邮件 ------------------ 发件人: "G Pavan Bhargav"<notifications@github.com>; 发送时间: 2020年6月3日(星期三) 下午2:34 收件人: "babloogpb1/Optic-disc-segmentation-using-pca-and-mathematical-morphology"<Optic-disc-segmentation-using-pca-and-mathematical-morphology@noreply.github.com>; 抄送: "1592404966"<1592404966@qq.com>;"Author"<author@noreply.github.com>; 主题: Re: [babloogpb1/Optic-disc-segmentation-using-pca-and-mathematical-morphology] OD position (#1)

Hii, The centre of od is not the brightest pixel location. If you refer to the paper, the authors actually mentioned a procedure to find the centroid of od using some distance function and a close hole operator. We didn't implement that part in the code, but we computed the centroid by considering the brightest region(od) and finding its centre.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe.

[babloogpb1]

yes, we've used open-cv watershed function instead of the one given in the paper.

[hahabingo]

Thank you, I tested  watershed.py and contour_only.py . I find that I get multiple centers and multiple coordinates, as shown in the pictures I sent. I am wondering if you've been doing any work in this field lately? Can you share the latest code with me? Thank you very much!

------------------ 原始邮件 ------------------ 发件人: "G Pavan Bhargav"<notifications@github.com>; 发送时间: 2020年6月3日(星期三) 下午2:53 收件人: "babloogpb1/Optic-disc-segmentation-using-pca-and-mathematical-morphology"<Optic-disc-segmentation-using-pca-and-mathematical-morphology@noreply.github.com>; 抄送: "1592404966"<1592404966@qq.com>;"Author"<author@noreply.github.com>; 主题: Re: [babloogpb1/Optic-disc-segmentation-using-pca-and-mathematical-morphology] OD position (#1)

yes, we've used open-cv watershed function instead of the one given in the paper.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe.

[babloogpb1]

oh, I'm afraid to say that I'm not working on it any more. The code in the repo is the latest one.

[hahabingo]

Glad to hear from you.  Let me introduce myself. I'm from Suzhou, China. I am a graduate student in my first year. And welcome to visit SuZhou.

I will continue to do this research in the future, and I hope that we can have more opportunities to communicate.

------------------ 原始邮件 ------------------ 发件人: "G Pavan Bhargav"<notifications@github.com>; 发送时间: 2020年6月3日(星期三) 晚上10:37 收件人: "babloogpb1/Optic-disc-segmentation-using-pca-and-mathematical-morphology"<Optic-disc-segmentation-using-pca-and-mathematical-morphology@noreply.github.com>; 抄送: "1592404966"<1592404966@qq.com>;"Author"<author@noreply.github.com>; 主题: Re: [babloogpb1/Optic-disc-segmentation-using-pca-and-mathematical-morphology] OD position (#1)

oh, I'm afraid to say that I'm not working on it any more. The code in the repo is the latest one.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe.

[babloogpb1]

Yeah. Glad to hear too. Thank you.