Video applications consume more energy on the Internet and can be accessed by electronic devices, due to an increase in the consumption of high-resolution and high-quality video content, presenting serious issues to delivery infrastructure that needs higher video compression technologies. The focus of this paper is to evaluate the quality of the most current codec, AV1, to its predecessor codec. The comparison was made experimentally at two video resolutions (1080p and 720p) by sampling video frames with various CRF/CQP values and testing several parameters analyses such as encoding duration, compression ratio, bit rate, Mean Square Error (MSE), and Peak Signal to Noise Ratio (PSNR). The AV1 codec is very great in terms of quality and file size, even though it is slower in terms of compression speed. The H.265/HEVC codec, on the other side, beats the other codec in terms of compression ratio. In conclusion, the H.265/HEVC codec is suggested as a material for obtaining a well compressed video with small file size and a short time.

Banyak sistem pemrosesan citra digital membutuhkan ekstraksi fitur di dalamnya. Salah satunya adalah ekstraksi foreground. Di dalam jurnal ini, kami mencoba melakukan ekstraksi foreground pada obyek daun melon dengan harapan hasil dari ekstraksi foreground dapat lebih lanjut dimanfaatkan terutama dalam proses pembuatan aplikasi yang berhubungan dengan daun melon, seperti misalnya pendeteksian dini terhadap penyakit daun melon. Dalam jurnal ini ekstraksi foreground dilakukan dengan bantuan algoritma GrabCut dengan bantuan deep neural network dan diaplikasikan sekaligus pada data obyek daun melon yang banyak. Hasilnya pada pengujian sebanyak 351 citra, ada 68% citra yang dapat diekstraksi citra daunya dengan sempurna.