Page 72 - 2023-Vol19-Issue2

68 | Odey & Marhoon

TABLE I.
Comparison between Our Model with Other Models

Ayegu¨l Uc¸ar et al Year Procedure Datasets Accuracy
2016 Caltech
Kai Hanet et al CNN with different layers Caltech pedestrians defection dataset 92.80+-0.43%
2017 Established a database consisting of defect area images
Stefan Schneider et al Faster RCNN 86.3%
Chandan G et al 2018 Reconyx camera trap self labeled gold standard snapshot Serengeti dataset
Xiangmo Zhao et al 2018 ResNet101 Camera 93.0%
Muhammad Faique Shakeel et al 2019 Faster RCNN Kitti dataset
Muhammad Azri Yahya et al 2019 Custom dataset 76.7%
2020 YOLO V2.0 Collected Data 99.0%
YANFEN LI et al Mobile Nets with SSD BDD100k 89.04%
Our model 2020 3d lider CNN Three collected datasets
2022 Our features dataset 78.18%
Camera CNN 84%
Novel deep based on CNN 71.43%
Lider YOLO v2.0
57.14%
SSD
52.7%
Modified YOLOv4
100%
Proposed hybrid model

national Conference on Software and Computer Appli- [10] H. Hakim and A. Fadhil, “Survey: Convolution neu-
cations, pp. 279–287, 2021. ral networks in object detection.,” Journal of Physics:
Conference Series., vol. 1804, no. 1, 2021.
[5] X. Ye and Q. Zhu, “Class-incremental learning based
on feature extraction of cnn with optimized softmax and [11] A. Uc¸ar, Y. Demir, and C. Gu¨zelis¸, “Moving towards in
one-class classifiers,” IEEE Acces, vol. 7, pp. 42024– object recognition with deep learning for autonomous
42031, 2019. driving applications,” in 2016 International Symposium
on INnovations in Intelligent SysTems and Applications
[6] M. Maggipinto, A. Beghi, S. McLoone, and G. A. Susto, (INISTA), pp. 1–5, IEEE, 2016.
“Deepvm: A deep learning-based approach with auto-
matic feature extraction for 2d input data virtual metrol- [12] K. Han, M. Sun, X. Zhou, G. Zhang, H. Dang, and
ogy,” Journal of Process Control, vol. 84, pp. 24–34, Z. Liu, “A new method in wheel hub surface defect de-
2019. tection: Object detection algorithm based on deep learn-
ing,” in 2017 International Conference on Advanced
[7] X. Li, W. Zhang, and Q. Ding, “Deep learning-based Mechatronic Systems (ICAMechS), pp. 335–338, IEEE,
remaining useful life estimation of bearings using multi- 2017.
scale feature extraction,” Reliability engineering & sys-
tem safety, vol. 182, pp. 208–218, 2019. [13] S. Schneider, G. W. Taylor, and S. Kremer, “Deep learn-
ing object detection methods for ecological camera trap
[8] Y. Xianjia, S. Salimpour, J. P. Queralta, and T. Wester- data,” in 2018 15th Conference on computer and robot
lund, “Analyzing general-purpose deep-learning detec- vision (CRV), pp. 321–328, IEEE, 2018.
tion and segmentation models with images from a lidar
as a camera sensor,” arXiv preprint arXiv:2203.04064, [14] G. Chandan, A. Jain, and H. Jain, “Real time object
2022. detection and tracking using deep learning and opencv,”
2018 International Conference on inventive research in
[9] M. Abdulla and A. Marhoon, “Agriculture based on computing applications (ICIRCA). IEEE, 2018.
internet of things and deep learning. iraqi journal for
electrical and electronic engineering 18.2 (2022): 1- [15] X. Zhao, P. Sun, Z. Xu, H. Min, and H. Yu, “Fusion of 3d
8.kim, cheol-jin, et al.end-to-end deep learning-based au- lidar and camera data for object detection in autonomous
tonomous driving control for high-speed environment.,” vehicle applications,” IEEE Sensors Journal, vol. 20,
The Journal of Supercomputing, vol. 78, no. 2, pp. 1961– no. 9, pp. 4901–4913, 2020.
1982, 2022.