Deteksi Objek Bergerak Pada Video Bawah Air Menggunakan Metode Frame Differencing
DOI:
https://doi.org/10.21776/jeeccis.v13i2.598Abstract
Abstract—Fish finder is an electronic device that operated on a ship that can measure the depth of sea. The principle work of the device is by measuring the depth of the ocean by pulse of vibration sound. There is a transducer that transmits pulse vibrations vertically to the bottom of the sea, then the transducer will receive the reflection of its. To increase modern fishing technology, a technology research using other methods are needed. One idea of modern fishing technology is object detection methods. There are several methods that can be used in the detection of moving objects. One of the method is the frame differencing method. This research designed the program of underwater object detection using frame differencing method by taking pictures with moving camera. The video was recorded using action camera. There are 3 kinds of data are processed according to the time, morning, noon, and night. This method compares two frames then the difference is considered the movement of an object. The effectiveness of this method is evaluated based on the value of recall and precision. The results showed that frame differencing method successfully detected the objects of the three videos with the highest value of recall is 73% and precision around 96% and execute duration in 0.36s. Index Terms— Object Detection, Underwater Detection, Frame Differencing.  Abstrak–- Salah satu gagasan mengenai teknologi penangkapan ikan modern adalah dengan menggunakan metode deteksi objek. Terdapat beberapa metode yang dapat digunakan dalam proses pendeteksian objek bergerak. Salah satu metode yang digunakan untuk mengidentifikasi obyek adalah metode frame differencing. Pada penelitian ini dirancang program deteksi obyek dalam air menggunakan metode frame differencing dengan pengambilan gambar dengan kamera bergerak. Penggunaan kamera bergerak memerlukan preprocessing pada video agar latar belakang menjadi seragam dan dapat dilakukan pembandingan antar frame. Pada penelitian ini data berupa video yang direkam menggunakan action cam. Penelitian ini menggunakan 3 buah data video yang diolah sesuai waktu pengambilan gambar, yaitu pagi, siang, dan malam. Metode yang diusulkan dilakukan dengan membandingkan dua buah frame yang yang berurutan dan selisihnya dianggap pergerakan dari sebuah objek. Hasil pembandingan diberi tanda untuk menyatakan objek yang dideteksi. Efektivitas metode ini dievaluasi berdasarkan nilai recall. Precision, dan waktu eksekusi. Hasil penelitian menunjukkan bahwa metode frame differencing berhasil mendeteksi objek dengan nilai recall tertinggi sebesar 73% dan precision tertinggi sebesar 96% serta waktu eksekusi tercepat 0.36s. Kata Kunci— Deteksi objek, frame differencing, action cam, recall, precision.ÂReferences
Amandeep, E.M.Goyal, “Moving object detection using background subtraction techniques,†International Journal Of Engineering Sciences & Research Technology, vol. 5, 2016, pp. 1-9.
M. R. Prabowo, N. Hudayani, S. Purwiyanti, S. R. Sulistiyanti, F. X. A. Setyawan, "A Moving Objects Detection in Underwater Video Using Subtraction of The Background Model," Proc. EECSI 2017, Yogyakarta, Indonesia, 19-21 September 2017.
Y. Yu, C. Zhou, L. Huang, Z. Yu, “A moving target detection algorithm based on the dynamic background,†International Conference on Computational Intelligence and Software Engineering, Wuhan – China, pp. 1-5, Dec. 2009.
M. R. Shortis, E. S. Harvey, "Design and calibration of an underwater stereo-video system for the monitoring of marine fauna populations," International Archives Photogrammetry and Remote Sensing, Vol. 32, No. 5, pp. 792-799, 1998.
M. S. Srividya, R. Hemavathy, G. Shobha, "Underwater video processing for detecting and tracking moving object," International Journal Of Engineering And Computer Science, Vol. 3, No. 5, Pp 5843-5847, May 2014.
C. Spampinato, Y. H. Chen-Burger, G. Nadarajan, and R. Fisher, “Detecting, tracking and counting fish in low quality unconstrained underwater videos,†Proc. 3rd Int. Conf. on Computer Vision Theory and Applications (VISAPP), vol. 2, pp. 514-519, 2008.
O. Ancha, "Speed algorithm for underwater moving objects detection based on image sequence," International Journal of Computer Science and Information Technologies, Vol. 7, No. 3, pp. 1032-1035, 2016.
K.D. Irianto, G. Ariyanto, D. Ary, "Motion Detection Using Opencv With Background Subtraction And Frame Differencing Technique," Simposium Nasional RAPI VIII 2009, pp. 74-81
F. X. A. Setyawan, J. K. Tan, H. Kim, S. Ishikawa, “Detecting Foreground Objects by Sequential Background Inference in a Video Captured by a Moving Camera,†Proceedings of the SICE Annual Conference, Nagoya - Japan, pp. 1699-1702, September 2013.
