Faster R-CNN for Detecting Regions in Human-Annotated Micrograph Images
Abstract
Object detection is a computer vision problem that involves classifying objects and identifying their location in a picture, video, or webcam feed. Applying deep learning techniques to this problem has led to huge contributions and attracted much research in this field. In this work, we use Faster RCNN to detect objects in micrograph images that have been manually annotated. We considered a Faster RCNN model because of its speed and its high accuracy of detection. The objective of this work is to detect objects in medical images and to evaluate the performance of Faster RCNN for medical images. We also compared the detecting highly overlapped objects and non-overlapping objects. Our experiment included 65 micrograph images with approximately 1000 objects, with findings that demonstrate that Faster RCNN obtained highly accurate results with mPA 76% for non overlapping objects.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
A faster R-CNN model using a dataset of symmetric objects improved accuracy rate of detecting defective and non-defective objects by increasing the training dataset to up to 400 images.
“The accuracy rate of detecting defective and non-defective objects is successfully improved, increasing the training dataset to up to 400 images of defective and non-defective objects.”
From Advanced Applications of Industrial Robotics: New Trends and Possibilities (2021) · cites this paper as [56]
Cites
ImageNet classification with deep convolutional neural networks
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017) · Communications of the ACM · doi:10.1145/3065386 · Cited by 112,155 · see citing works
Soft-NMS — Improving Object Detection with One Line of Code
Bodla, N., Singh, B., Chellappa, R., & Davis, L. S. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.593 · Cited by 2,274 · cited 1× · see citing works
“It worthwhile to mention that Non Maximum Suppression implies that for images that contain more than one overlapping objects, the detector will only consider the highest IOU to be TP and it will discard the rest of boxes to be FP [14] [2].”
Background
Flagged first, then by citations
Cited by
Sorted by citations
Advanced Applications of Industrial Robotics: New Trends and Possibilities
Dzedzickis, A., Subačiūtė-Žemaitienė, J., Šutinys, E., Samukaitė-Bubnienė, U., & Bučinskas, V. (2021) · Applied Sciences · doi:10.3390/app12010135 · Cited by 228 · see citing works
- Authors
- Albinali, H., & Alzahrani, F. A.
- Affiliations
- King Fahd University of Petroleum and Minerals (SA)
- Venue
- 2021 International Conference of Women in Data Science at Taif University (WiDSTaif )
- Publisher
- IEEE
- Published
- 2021-03-30
- Volume
- 26
- Pages
- 1-6
- OpenAlex
- W3170035274
- Cited by
- 2
- References
- 2
- Citations by year
- 2024: 1 · 2021: 1
- FWCI
- 0.16
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- International Conference of Women in Data Science at Taif University (WiDSTaif) (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Biophysics › Cell Image Analysis Techniques
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Medical Image Segmentation Techniques
- Keywords
- Faster R-CNN, object detection, medical image analysis, overlapping object detection, detection accuracy evaluation
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
