What is Deep learning approach to count the number of Diamonds in an image?

Question

I am working on a project which involves counting the number of diamonds in the provided image. I have a set of images and a VIA annotated .json file which has all the annotations. How do I proceed with this problem.

what approach to start i.e Object detection vs Instance segmentation ?
what models to choose from MaskRCNN,U-Net, etc (Please suggest) ?

Please guide. Thanks

AIFahim · Accepted Answer · 2021-01-28 02:37:11Z

1

I say that go with Object Detection. Because you already have annoted images. And segmentation applied to locate objects and boundaries (lines, curves, etc.) in images. Ref : ][1] . In Recent days EfficientDet Works very well in object detection. And for implementation look here. Best of luck.

answered Jan 28 at 2:37

AIFahim

2551 gold badge1 silver badge12 bronze badges

1

$\begingroup$ Yes obviously if you detect the object then you can apply segmentation on them. Look at this paper (ceur-ws.org/Vol-2595/endoCV2020_paper_id_11.pdf). In segmentation is detect the belonging pixels of the objects. Which can be done after the detection cause after finds the location we do segmentations. So that detection first is good for you because you have labelled images. Best of luck. $\endgroup$ – AIFahim Jan 28 at 5:00
1

$\begingroup$ Look at this code for details explanation (kaggle.com/shonenkov/training-efficientdet ). This code from Global Wheat Detection kaggle (kaggle.com/c/global-wheat-detection/discussion/152368). Another implementation I already mentioned (github.com/rwightman/…). Hope this help. If you find useful and it solves your problem upvote and mark inspire me a lot. $\endgroup$ – AIFahim Jan 28 at 12:11
1

$\begingroup$ Thanks a lot. Any time I will happy and appreciate then. Recently Kaggle launch a competition I feel similarly with your problem. Check this out (kaggle.com/c/hpa-single-cell-image-classification) . And you will also find me in Kaggle (kaggle.com/aifahim ) anytime. Best of luck . $\endgroup$ – AIFahim Jan 29 at 6:19
1

$\begingroup$ If you follow the discussion thread and notebooks of this competition you will be benefited. $\endgroup$ – AIFahim Jan 29 at 6:22
1

$\begingroup$ Discussion Thread (kaggle.com/c/hpa-single-cell-image-classification/discussion) and Notebooks thread (kaggle.com/c/hpa-single-cell-image-classification/notebooks). In Kaggle competitions you always find the state of art updated solutions. And this competition is similar with your problems which means if you follow this , it will be cheery on the top for you :) $\endgroup$ – AIFahim Jan 29 at 6:33

| Show 8 more comments

Stack Exchange Network

current community

your communities

more stack exchange communities

What is Deep learning approach to count the number of Diamonds in an image?

1 Answer 1

Your Answer

Not the answer you're looking for? Browse other questions tagged deep-learning cnn image-segmentation vgg16 or ask your own question.

Hot Network Questions

What is Deep learning approach to count the number of Diamonds in an image?

1 Answer 1

Your Answer

Sign up or log in

Post as a guest

Not the answer you're looking for? Browse other questions tagged deep-learning cnn image-segmentation vgg16 or ask your own question.

Related

Hot Network Questions