International Peer-Reviewed Journal•Open Access•ISSN 2456-8880
[email protected]•+91-7433024337

Home / Current Issue / Paper 1725253

1725253 Vol 10 · Issue 5 Download Paper

Enhanced Small Object Detection in High – Resolution Images Using CNN-Boosted YOLOv5

Dr. S. Lakshmi Narayanan Dharsan B J Kailash S B Chinnasamy P

Subject area: Science,Engineering and Technology  ·  Area of research: Image Processing, Machine and Deep Learning

Abstract

Small object detection remains difficult in high-resolution images because of the limited spatial information, occlusion, and complicated background scenes. This paper uses a CNN-Boosted YOLOv5 framework to construct an improved small object detection framework that is tailored to improve feature representation when using high-resolution visual images. The proposed method incorporates a highly enhanced CNN-based feature extractor within the YOLOv5 detection system, which allows it to maintain finer-grained spatial information that is required to detect small, distant, and partially covered objects. The experimental outcomes reveal that there are observed improvements in the accuracy of localization, detection confidence and robustness in different real-life conditions. The model can be used in adverse environments, like traffic, crowded areas, and cluttered backgrounds, and it can be used to infer in real-time. The CNN-Boosted YOLOv5 network is appropriate in executing real-time systems by applying an effective trade-off between the accuracy of detection and the efficiency of computation. On the whole, the experiment proves that the performance of small objects detection can be improved by much when the features in the YOLOv5 are enhanced which means that the given approach can be widely applied to autonomous driving, intelligent transportation systems, surveillance, and smart city solutions.

Keywords

Small object detection, high-resolution images, YOLOv5, CNN boosting, deep learning, multi-scale feature extraction, object detection, real-time detection, convolutional neural networks, computer vision

References

[1] Feng D, Harakeh A, Waslander SL, Dietmayer K. A review and comparative study on probabilistic object detection in autonomous driving. IEEE Trans Intell Transp Syst. 2022;23(8):Syst.

[2] Kaur A, Singh Y, Neeru N, Kaur L, Singh A. A survey on deep learning approaches to medical images and a systematic look up into real-time object detection. Arch Comput Methods Eng. 2022;29(4):2071–2111.

[3] Hayat M, Aramvith S, Achakulvisut T. Combined channel and spatial attention-based stereo endoscopic image super-resolution. In: Proceedings of the IEEE Region 10 Conference (TENCON); 2023. p. 920–925.

[4] Hayat M, Aramvith S, Achakulvisut T. SEGSRNet for stereo-endoscopic image super-resolution and surgical instrument segmentation. arXiv:2404.13330 [Preprint]. 2024.

[5] Gomaa A, Minematsu T, Abdelwahab MM, Abo-Zahhad M, Taniguchi RI. Faster CNN-based vehicle detection and counting strategy for fixed camera scenes. Multimedia Tools Appl. 2022;81(18):25443–25471.

[6] Salem M, Gomaa A, Tsurusaki N. Detection of earthquake-induced building damages using remote sensing data and deep learning: A case study of Mashiki town, Japan. In: Proceedings of the IEEE International Geoscience and Remote Sensing Symposium (IGARSS); 2023. p. 2350–2353.

[7] Zhai S, Shang D, Wang S, Dong S. DF-SSD: An improved SSD object detection algorithm based on DenseNet and feature fusion. IEEE Access. 2020;8:24344–24357.

[8] Baidya R, Jeong H. YOLOv5 with ConvMixer prediction heads for precise object detection in drone imagery. Sensors. 2022;22(21):8424.

[9] Jocher G, Stoken A, Borovec J, Chaurasia A, et al. ultralytics/yolov5: v5.0 – YOLOv5-P6 1280 models, AWS, Supervise.ly and YouTube integrations [Software]. Zenodo; 2021. https://doi.org/10.5281/zenodo.4679653

[10] Bochkovskiy A, Wang CY, Liao HYM. YOLOv4: Optimal speed and accuracy of object detection. arXiv:2004.10934 [Preprint]. 2020.

[11] Akyon FC, Altinuc SO, Temizel A. Slicing aided hyper inference and fine-tuning for small object detection. In: 2022 IEEE International Conference on Image Processing (ICIP); 2022. p. 966–970. https://doi.org/10.1109/ICIP46576.2022.9897990

[12] Dosovitskiy A, Beyer L, Kolesnikov A, Weissenborn D, Zhai X, Unterthiner T, et al. An image is worth 16x16 words: Transformers for image recognition at scale. In: International Conference on Learning Representations; 2021.

[13] David R, Duke J, Jain A, Janapa Reddi V, Jeffries N, Li J, et al. TensorFlow Lite Micro: Embedded machine learning for TinyML systems. In: Proceedings of Machine Learning and Systems. Vol. 3; 2021. p. 800–811.

[14] Moosmann J, Giordano M, Vogt C, Magno M. TinyissimoYOLO: A quantized, low-memory footprint, TinyML object detection network for low power microcontrollers. arXiv:2306.00001 [Preprint]. 2023.

[15] Cerutti G, Andri R, Cavigelli L, Farella E, Magno M, Benini L. Sound event detection with binary neural networks on tightly power-constrained IoT devices. In: Proceedings of the ACM/IEEE International Symposium on Low Power Electronics and Design; 2020. p. 19–24.

How to cite this paper

Dr. S. Lakshmi Narayanan, Dharsan B J, Kailash S B, Chinnasamy P "Enhanced Small Object Detection in High – Resolution Images Using CNN-Boosted YOLOv5" Iconic Research And Engineering Journals Volume 10 Issue 5 2026 Page 3507-3521
Dr. S. Lakshmi Narayanan, Dharsan B J, Kailash S B, Chinnasamy P "Enhanced Small Object Detection in High – Resolution Images Using CNN-Boosted YOLOv5" Iconic Research And Engineering Journals, vol. 10, no. 5, Nov. 2026
Dr. S. Lakshmi Narayanan, Dharsan B J, Kailash S B, Chinnasamy P (2026). Enhanced Small Object Detection in High – Resolution Images Using CNN-Boosted YOLOv5. Iconic Research And Engineering Journals, 10(5).
Dr. S. Lakshmi Narayanan, Dharsan B J, Kailash S B, Chinnasamy P "Enhanced Small Object Detection in High – Resolution Images Using CNN-Boosted YOLOv5" Iconic Research And Engineering Journals, vol. 10, no. 5, Nov. 2026.
@article{1725253,
      author = {Dr. S. Lakshmi Narayanan, Dharsan B J, Kailash S B, Chinnasamy P},
      title = {Enhanced Small Object Detection in High – Resolution Images Using CNN-Boosted YOLOv5},
      journal = {Iconic Research And Engineering Journals},
      year = {2026},
      volume = {10},
      number = {5},
      pages = {3507-3521},
      issn = {2456-8880},
      url = {https://www.irejournals.com/formatedpaper/1725253.pdf},
      abstract = {Small object detection remains difficult in high-resolution images because of the limited spatial information, occlusion, and complicated background scenes. This paper uses a CNN-Boosted YOLOv5 framework to construct an improved small object detection framework that is tailored to improve feature representation when using high-resolution visual images. The proposed method incorporates a highly enhanced CNN-based feature extractor within the YOLOv5 detection system, which allows it to maintain finer-grained spatial information that is required to detect small, distant, and partially covered objects. The experimental outcomes reveal that there are observed improvements in the accuracy of localization, detection confidence and robustness in different real-life conditions. The model can be used in adverse environments, like traffic, crowded areas, and cluttered backgrounds, and it can be used to infer in real-time. The CNN-Boosted YOLOv5 network is appropriate in executing real-time systems by applying an effective trade-off between the accuracy of detection and the efficiency of computation. On the whole, the experiment proves that the performance of small objects detection can be improved by much when the features in the YOLOv5 are enhanced which means that the given approach can be widely applied to autonomous driving, intelligent transportation systems, surveillance, and smart city solutions.},
      keywords = {Small object detection, high-resolution images, YOLOv5, CNN boosting, deep learning, multi-scale feature extraction, object detection, real-time detection, convolutional neural networks, computer vision},
      month = {November},
  }