ELOS 2022

Yilmaz Eren Bolukbasi

Yilmaz Eren Bolukbasi, Speaker at Optics conferences
Aselsan, Gazi University, Turkey
Title : FPGA implementation of laser spot detection algorithm

Abstract:

It is aimed to effectively detect the laser spot which are frequently used in defense industry systems, even in adverse weather conditions such as fog and rain, on the FPGA by eliminating moving elements. In this study which is the subject of this article, three video frames were taken consecutively, and it was accepted that the first of them carried a laser spot and the second and third did not. In this way, moving elements such as birds, airplanes, helicopters, unmanned aerial vehicles, people or animals which have an image close to the laser spot, will be separated from the laser and more efficient results will be obtained in foggy and rainy weather conditions.
In order to get better results than the method used in previous studies, three video frames taken will be run on separated flows from the two branches and these branches will be combined in the last step of the algorithm. In the first flow, the difference matrix will be created by subtracting the next video frame without laser spot from the photo frame with laser spot, and non-linear median filtering method will be applied on this matrix in order to reduce the salt and pepper noise in the image at certain dimensions. Afterwards, it is aimed to soften the image in proportion to the selected mask value by outputting low-frequency video signals with the average filter from the convolution filters. In the third stage, after a threshold value calculation is made, the pixels above this value will be kept on the flow. Operations in the first flow will be completed to these in order to preserve the laser spot. In the second flow, which will continue in parallel with the first flow, the difference matrix of two consecutive laser unspotted frames will be created and the same methods with the first flow will be applied. Unlike the first flow, the second flow will be eroded with image erosion after the threshold filtering step. After the image dilation process, which is one of the morphological image processing methods, the image will be extended. In the first flow, in order to remove the moving elements in the image known to be lasered, after this process, morphological image processing techniques will be applied again in the second flow. Then, information from the matrices from the first and second flows will be combined to determine whether there is laser spot information and in which row and column it is located.
Laser target marking device is capable of making laser pulses at determined frequencies, and it is guaranteed by the system requirements that a second laser pulse cannot be obtained in the time required for taking three video frames. Morphological image processing methods applied to achieve the result will be explained and the performance of the new algorithm will be measured by comparing the simulation results of the codes written in the VHDL language with the data obtained in the MATLAB calculation tool.

Audience Take Away:

  •  Audience will learn alternative image processing techniques in laser spot detection applications and will be able to try different algorithms in new studies by seeing examples in the VHDL language created to implement them on hardware.
  • They will have a vision to produce similar algorithms applied in real-time defense industry systems in civilian areas such as medical and transportation.
  • Median filter, mean filter, threshold filter, erosion and dilation applications can be virtually seen on simulation results.
     

Biography:

Yilmaz Eren Bolukbasi was born in 1989 in Yozgat. He graduated from Ankara Science High School and Yeditepe University Electronics Engineering with a scholarship. He still works as a digital design engineer at ASELSAN Microelectronics, Guidance & Electro-Optics Business Sector, which is the world 48th biggest defence industry company. He is a graduate student in Gazi University Electrical and Electronics Engineering Department and a Konya Necmettin Erbakan University Faculty of Law undergraduate student.Yılmaz Eren Bölükbaşı was born in 1989 in Yozgat. He graduated from Ankara Science High School and Yeditepe University Electronics Engineering with a scholarship. He still works as a digital design engineer at ASELSAN Microelectronics, Guidance & Electro-Optics Business Sector, which is the world 48th biggest defence industry company. He is a graduate student in Gazi University Electrical and Electronics Engineering Department and a Konya Necmettin Erbakan University Faculty of Law undergraduate student.

Watsapp