Cite Article

Reference Class-Based Improvement of Object Detection Accuracy

Choose citation format

BibTeX

@article{IJASEIT12792,
   author = {Raegeun Park and Jaechoon Jo},
   title = {Reference Class-Based Improvement of Object Detection Accuracy},
   journal = {International Journal on Advanced Science, Engineering and Information Technology},
   volume = {10},
   number = {4},
   year = {2020},
   pages = {1526--1535},
   keywords = {reference class; target class; FP case; association; improvement of accuracy performance.},
   abstract = {

To date, the Frames Per Second (FPS) and accuracy of object detection based on deep learning have made rapid progress. However, the accuracy is limited by issues such as false positive (FP) cases. FP cases can trigger malfunctions in applications requiring high accuracy, such as in autonomous vehicles, where it is essential to ensure driver safety when malfunctions occur. To reduce the occurrences of FP cases, we conducted an experiment to derive the association by separately detecting a highly relevant element called a reference class, in addition to the target class to be detected. To measure the association, we obtained the integrated association by first finding the associations between the bounding boxes of the target and reference classes. Then we generated a reference class-based model by applying the integrated association to a trained model. The reference class-based model achieved approximately 15% higher accuracy than the trained model at iteration 1,000. Besides, the proposed model reduced the FP cases to approximately half of the 18.964% in the conventional method; the FP reduction through an increase in iteration was only 11.008%. The reference class can be applied in various fields, such as security and autonomous vehicle technology. It can be used to reduce the FP cases and improve the accuracy performance limits in object detection. Furthermore, it is possible to reduce the cost of reinforcing the training dataset and using high-performance hardware, and the time cost of increasing training numbers.

},    issn = {2088-5334},    publisher = {INSIGHT - Indonesian Society for Knowledge and Human Development},    url = {http://ijaseit.insightsociety.org/index.php?option=com_content&view=article&id=9&Itemid=1&article_id=12792},    doi = {10.18517/ijaseit.10.4.12792} }

EndNote

%A Park, Raegeun
%A Jo, Jaechoon
%D 2020
%T Reference Class-Based Improvement of Object Detection Accuracy
%B 2020
%9 reference class; target class; FP case; association; improvement of accuracy performance.
%! Reference Class-Based Improvement of Object Detection Accuracy
%K reference class; target class; FP case; association; improvement of accuracy performance.
%X 

To date, the Frames Per Second (FPS) and accuracy of object detection based on deep learning have made rapid progress. However, the accuracy is limited by issues such as false positive (FP) cases. FP cases can trigger malfunctions in applications requiring high accuracy, such as in autonomous vehicles, where it is essential to ensure driver safety when malfunctions occur. To reduce the occurrences of FP cases, we conducted an experiment to derive the association by separately detecting a highly relevant element called a reference class, in addition to the target class to be detected. To measure the association, we obtained the integrated association by first finding the associations between the bounding boxes of the target and reference classes. Then we generated a reference class-based model by applying the integrated association to a trained model. The reference class-based model achieved approximately 15% higher accuracy than the trained model at iteration 1,000. Besides, the proposed model reduced the FP cases to approximately half of the 18.964% in the conventional method; the FP reduction through an increase in iteration was only 11.008%. The reference class can be applied in various fields, such as security and autonomous vehicle technology. It can be used to reduce the FP cases and improve the accuracy performance limits in object detection. Furthermore, it is possible to reduce the cost of reinforcing the training dataset and using high-performance hardware, and the time cost of increasing training numbers.

%U http://ijaseit.insightsociety.org/index.php?option=com_content&view=article&id=9&Itemid=1&article_id=12792 %R doi:10.18517/ijaseit.10.4.12792 %J International Journal on Advanced Science, Engineering and Information Technology %V 10 %N 4 %@ 2088-5334

IEEE

Raegeun Park and Jaechoon Jo,"Reference Class-Based Improvement of Object Detection Accuracy," International Journal on Advanced Science, Engineering and Information Technology, vol. 10, no. 4, pp. 1526-1535, 2020. [Online]. Available: http://dx.doi.org/10.18517/ijaseit.10.4.12792.

RefMan/ProCite (RIS)

TY  - JOUR
AU  - Park, Raegeun
AU  - Jo, Jaechoon
PY  - 2020
TI  - Reference Class-Based Improvement of Object Detection Accuracy
JF  - International Journal on Advanced Science, Engineering and Information Technology; Vol. 10 (2020) No. 4
Y2  - 2020
SP  - 1526
EP  - 1535
SN  - 2088-5334
PB  - INSIGHT - Indonesian Society for Knowledge and Human Development
KW  - reference class; target class; FP case; association; improvement of accuracy performance.
N2  - 

To date, the Frames Per Second (FPS) and accuracy of object detection based on deep learning have made rapid progress. However, the accuracy is limited by issues such as false positive (FP) cases. FP cases can trigger malfunctions in applications requiring high accuracy, such as in autonomous vehicles, where it is essential to ensure driver safety when malfunctions occur. To reduce the occurrences of FP cases, we conducted an experiment to derive the association by separately detecting a highly relevant element called a reference class, in addition to the target class to be detected. To measure the association, we obtained the integrated association by first finding the associations between the bounding boxes of the target and reference classes. Then we generated a reference class-based model by applying the integrated association to a trained model. The reference class-based model achieved approximately 15% higher accuracy than the trained model at iteration 1,000. Besides, the proposed model reduced the FP cases to approximately half of the 18.964% in the conventional method; the FP reduction through an increase in iteration was only 11.008%. The reference class can be applied in various fields, such as security and autonomous vehicle technology. It can be used to reduce the FP cases and improve the accuracy performance limits in object detection. Furthermore, it is possible to reduce the cost of reinforcing the training dataset and using high-performance hardware, and the time cost of increasing training numbers.

UR - http://ijaseit.insightsociety.org/index.php?option=com_content&view=article&id=9&Itemid=1&article_id=12792 DO - 10.18517/ijaseit.10.4.12792

RefWorks

RT Journal Article
ID 12792
A1 Park, Raegeun
A1 Jo, Jaechoon
T1 Reference Class-Based Improvement of Object Detection Accuracy
JF International Journal on Advanced Science, Engineering and Information Technology
VO 10
IS 4
YR 2020
SP 1526
OP 1535
SN 2088-5334
PB INSIGHT - Indonesian Society for Knowledge and Human Development
K1 reference class; target class; FP case; association; improvement of accuracy performance.
AB 

To date, the Frames Per Second (FPS) and accuracy of object detection based on deep learning have made rapid progress. However, the accuracy is limited by issues such as false positive (FP) cases. FP cases can trigger malfunctions in applications requiring high accuracy, such as in autonomous vehicles, where it is essential to ensure driver safety when malfunctions occur. To reduce the occurrences of FP cases, we conducted an experiment to derive the association by separately detecting a highly relevant element called a reference class, in addition to the target class to be detected. To measure the association, we obtained the integrated association by first finding the associations between the bounding boxes of the target and reference classes. Then we generated a reference class-based model by applying the integrated association to a trained model. The reference class-based model achieved approximately 15% higher accuracy than the trained model at iteration 1,000. Besides, the proposed model reduced the FP cases to approximately half of the 18.964% in the conventional method; the FP reduction through an increase in iteration was only 11.008%. The reference class can be applied in various fields, such as security and autonomous vehicle technology. It can be used to reduce the FP cases and improve the accuracy performance limits in object detection. Furthermore, it is possible to reduce the cost of reinforcing the training dataset and using high-performance hardware, and the time cost of increasing training numbers.

LK http://ijaseit.insightsociety.org/index.php?option=com_content&view=article&id=9&Itemid=1&article_id=12792 DO - 10.18517/ijaseit.10.4.12792