Characteristics of Stereo Imaging using Non-Identical Cameras for Object Detection

##plugins.themes.bootstrap3.article.main##

  •   Ahmed M. D. E. Hassanein

  •   Amira H. N. AboElanen

  •   Salma Ahmed H. Z.

Abstract

The fear from the continuous spreading of the Covid-19 pandemic had put lot of restrictions on the movement of goods around the world. In Egypt, the importing of goods especially electronic products from many countries including China was crucial to the research and educational purposes. The restrictions had stopped the importing of many electronic devices from China including cameras. Object detection and identification were among the hot topics of research in our university which depended mainly on imported cameras. In this paper we tackle the problem of setting up stereo cameras using old non identical cameras to do object detection. The selection of the cameras was not optional since we had to use what we found in our old laptops. OpenCV and Python programming commands were used to set the two cameras to obtain equally clear images as much as possible. A disparity map was then calculated using openCV and its accuracy was then discussed. Accuracy was dependable on the sharpness of the cameras used, Gamma parameter, number of pixels per image and matching algorithm to match the two images obtained using the stereo cameras.


Keywords: Disparity, Stereo Camera, OpenCV, Object Detection, Identification

References

Research Note on Impacts of COVID-19 Pandemic on Egypt’s Economy; March 2020; American Chamber of Commerce in Egypt (AmCham Egypt).

Christodoulou, Lakis. 3D Stereo Vision Camera-sensors, Advancements, and Technologies, 2013.

Object-Centric Stereo Matching for 3D Object Detection; Alex D. Pon and Jason Ku and Chengyao Li and Steven L. Waslander; arXiv:1909.07566v2 [cs.CV1] 10 Mar 2020.

https://www.jetbrains.com/pycharm/download/#section=windows.

https://opencv-python-tutroals.readthedocs.io/en/latest/py_tutorials/py_calib3d/py_depthmap/py_depthmap.html.

Distance measuring based on stereoscopic pictures, Jernej Mrovlje and Damir Vrancic, 9th International PhD Workshop on Systems and Control: Young Generation Viewpoint 1-3. October 2008, Izola, Slovenia.

DU LL, Wang LM, Zhuang Z. [Measurement and analysis of human head-face dimensions1]. Zhonghua Lao Dong Wei Sheng Zhi Ye Bing Za Zhi. 2008 May;26(5) 266-270. PMID: 18727867.

How Many Pixels Do We Need to See Things? Yang Cai, P.M.A. Sloot et al. (Eds.): ICCS 2003, LNCS 2659, pp. 1064–1073, 2003. Springer-Verlag Berlin Heidelberg 2003.

https://docs.opencv.org/3.4/d2/d85/classcv_1_1StereoSGBM.html.

On the Importance of Stereo for Accurate Depth Estimation: An Efficient Semi-Supervised Deep Neural Network Approach; Nikolai Smolyanskiy Alexey Kamenev Stan Birchfield; arXiv:1803.09719v4 [cs.CV1] 8 Jul 2020.

Johnston KJ, Hammond G, Meyers DJ, Joynt Maddox KE. Association of race and ethnicity and medicare program type with ambulatory care access and quality measures. JAMA. 2021 Aug 17;326(7):628-636. doi: 10.1001/jama.2021.10413.

Naderi SH, Bestwick JP, Wald DS. Adherence to drugs that prevent cardiovascular disease: meta-analysis on 376,162 patients. Am J Med. 2012; 125(9): 882–887.e1.

Downloads

Download data is not yet available.

##plugins.themes.bootstrap3.article.details##

How to Cite
[1]
Hassanein, A.M.D.E., AboElanen, A.H.N. and Ahmed H. Z., S. 2021. Characteristics of Stereo Imaging using Non-Identical Cameras for Object Detection. European Journal of Engineering and Technology Research. 6, 7 (Dec. 2021), 122–126. DOI:https://doi.org/10.24018/ejeng.2021.6.7.2670.