Small Object Detection via Pixel Level Balancing With Applications to Blood Cell Detection
Object detection technology has been widely used in medical field, such as detecting the images of blood cell to count the changes and distribution for assisting the diagnosis of diseases. However, detecting small objects is one of the most challenging and important problems especially in medical sc...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2022-06-01
|
| Series: | Frontiers in Physiology |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fphys.2022.911297/full |
| _version_ | 1811245275051720704 |
|---|---|
| author | Bin Hu Yang Liu Yang Liu Pengzhi Chu Minglei Tong Qingjie Kong |
| author_facet | Bin Hu Yang Liu Yang Liu Pengzhi Chu Minglei Tong Qingjie Kong |
| author_sort | Bin Hu |
| collection | DOAJ |
| description | Object detection technology has been widely used in medical field, such as detecting the images of blood cell to count the changes and distribution for assisting the diagnosis of diseases. However, detecting small objects is one of the most challenging and important problems especially in medical scenarios. Most of the objects in medical images are very small but influential. Improving the detection performance of small objects is a very meaningful topic for medical detection. Current researches mainly focus on the extraction of small object features and data augmentation for small object samples, all of these researches focus on extracting the feature space of small objects better. However, in the training process of a detection model, objects of different sizes are mixed together, which may interfere with each other and affect the performance of small object detection. In this paper, we propose a method called pixel level balancing (PLB), which takes into account the number of pixels contained in the detection box as an impact factor to characterize the size of the inspected objects, and uses this as an impact factor. The training loss of each object of different size is adjusted by a weight dynamically, so as to improve the accuracy of small object detection. Finally, through experiments, we demonstrate that the size of objects in object detection interfere with each other. So that we can improve the accuracy of small object detection through PLB operation. This method can perform well with blood cell detection in our experiments. |
| first_indexed | 2024-04-12T14:38:16Z |
| format | Article |
| id | doaj.art-9fe03e703c0b49e9a061470d19ebbb74 |
| institution | Directory Open Access Journal |
| issn | 1664-042X |
| language | English |
| last_indexed | 2024-04-12T14:38:16Z |
| publishDate | 2022-06-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Physiology |
| spelling | doaj.art-9fe03e703c0b49e9a061470d19ebbb742022-12-22T03:29:00ZengFrontiers Media S.A.Frontiers in Physiology1664-042X2022-06-011310.3389/fphys.2022.911297911297Small Object Detection via Pixel Level Balancing With Applications to Blood Cell DetectionBin Hu0Yang Liu1Yang Liu2Pengzhi Chu3Minglei Tong4Qingjie Kong5Department of Compute Science and Engineering, School of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University, Shanghai, ChinaDepartment of Dermatology, Shanghai Ninth People’s Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai, ChinaDepartment of Laser and Aesthetic Medicine, Shanghai Ninth People’s Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai, ChinaDepartment of Compute Science and Engineering, School of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University, Shanghai, ChinaCollege of Electronics and Information Engineering, Shanghai University of Electric Power, Shanghai, ChinaRiseye Research, Riseye Intelligent Technology (Shanghai) Co., Ltd., Shanghai, ChinaObject detection technology has been widely used in medical field, such as detecting the images of blood cell to count the changes and distribution for assisting the diagnosis of diseases. However, detecting small objects is one of the most challenging and important problems especially in medical scenarios. Most of the objects in medical images are very small but influential. Improving the detection performance of small objects is a very meaningful topic for medical detection. Current researches mainly focus on the extraction of small object features and data augmentation for small object samples, all of these researches focus on extracting the feature space of small objects better. However, in the training process of a detection model, objects of different sizes are mixed together, which may interfere with each other and affect the performance of small object detection. In this paper, we propose a method called pixel level balancing (PLB), which takes into account the number of pixels contained in the detection box as an impact factor to characterize the size of the inspected objects, and uses this as an impact factor. The training loss of each object of different size is adjusted by a weight dynamically, so as to improve the accuracy of small object detection. Finally, through experiments, we demonstrate that the size of objects in object detection interfere with each other. So that we can improve the accuracy of small object detection through PLB operation. This method can perform well with blood cell detection in our experiments.https://www.frontiersin.org/articles/10.3389/fphys.2022.911297/fullmedical image detectionobject detectionsmall objectpixel level balanceblood cell detection |
| spellingShingle | Bin Hu Yang Liu Yang Liu Pengzhi Chu Minglei Tong Qingjie Kong Small Object Detection via Pixel Level Balancing With Applications to Blood Cell Detection Frontiers in Physiology medical image detection object detection small object pixel level balance blood cell detection |
| title | Small Object Detection via Pixel Level Balancing With Applications to Blood Cell Detection |
| title_full | Small Object Detection via Pixel Level Balancing With Applications to Blood Cell Detection |
| title_fullStr | Small Object Detection via Pixel Level Balancing With Applications to Blood Cell Detection |
| title_full_unstemmed | Small Object Detection via Pixel Level Balancing With Applications to Blood Cell Detection |
| title_short | Small Object Detection via Pixel Level Balancing With Applications to Blood Cell Detection |
| title_sort | small object detection via pixel level balancing with applications to blood cell detection |
| topic | medical image detection object detection small object pixel level balance blood cell detection |
| url | https://www.frontiersin.org/articles/10.3389/fphys.2022.911297/full |
| work_keys_str_mv | AT binhu smallobjectdetectionviapixellevelbalancingwithapplicationstobloodcelldetection AT yangliu smallobjectdetectionviapixellevelbalancingwithapplicationstobloodcelldetection AT yangliu smallobjectdetectionviapixellevelbalancingwithapplicationstobloodcelldetection AT pengzhichu smallobjectdetectionviapixellevelbalancingwithapplicationstobloodcelldetection AT mingleitong smallobjectdetectionviapixellevelbalancingwithapplicationstobloodcelldetection AT qingjiekong smallobjectdetectionviapixellevelbalancingwithapplicationstobloodcelldetection |