Image data compression using fast Fourier transform (FFT) technique for wireless sensor network

• Main Authors: Haron, M. H., Isa, M. N. Md, Ahmad, M. I., Arshad, M. A. Mohamed, Jambek, A. B., Naziri, S. Z. M, Hussin, R., Ismail, R. C., Harun, A., Mohyar, S. N.
• Format: Conference Proceeding
• Language: English
• Subjects: Algorithms; Compression ratio; Data compression; Discrete cosine transform; Fast Fourier transformations; Fourier transforms; Image compression; Image quality; Signal to noise ratio; Snails; Wireless networks; Wireless sensor networks
• ISSN: 0094-243X; 1551-7616
• DOI: 10.1063/5.0195298

Agricultural settings present unique challenges for the transmission of huge amounts of images over long-range wireless networks. It is challenging to remotely gather data for transmission over a wireless network in research areas due to a lack of basic amenities like internet connections, especially in distant agricultural areas. In this research, the Fast Fourier Transform (FFT) method was used in conjunction with the Discrete Cosine Transform (DCT) method of image compression to achieve a higher compression ratio. In order for a Wireless Sensor Network (WSN) to provide compressed image data to a wireless based station, a LoRaWAN network has been identified. Low-power LoRaWAN networks may regularly transmit compressed images from an agricultural region to a monitoring system up to 15 km away. Images of golden apple snails were collected for this study from a variety of sources. The procedure was coded in MATLAB so that it could be run with input images being judged by the created algorithm. The input images can be compressed with a range of compression ratios (CR) from 3.00 to 50.00, as shown by the simulation results. Compressed image quality is measured not only by the above-mentioned criteria, but also by Mean Square Error (MSE) and Peak Signal to Noise Ratio (PSNR). According to the numbers, the best achievable compression ratio is 49.04, with an MSE of 172.72 and a PSNR of 25.75 at its highest.