현대 사회에서 첨단 기술의 발전은 예술의 양상을 빠르게 변화시키고 있다. 생성형 인공지능 기술의 발전은 인류의 전유물이라 여겨왔던 창작 행위에 기존에는 보지 못했던 색다른 표현을 제공했다. 또한, 전통적 기법 을 고수하던 예술인들에게 새로운 예술 창작의 발현 방식을 제시하였다. 하지만 나날이 발전하는 인공지능 기술에 비해, 이를 활용한 구체적인 미디어아트 제작 과정 연구는 아직 국내에서는 미비하다. 본 논문은 인 공지능 기술을 활용한 미디어아트 해외 사례를 탐구한다. 그리고 Text-to-Image 인공지능 생성 모델과 게임 엔진 Unreal Engine 5를 이용하여, 국내에 자리 잡지 않은 생성형 인공지능을 활용한 작품 제작 방법론과 창 작자들에게 인공지능 이미지 생성 모델의 확장성을 제시한다.

Cereal seeds, sorghum, foxtail millet, hog millet, adlay, and corn are traditionally used as health assistant as well as energy supplying food in Korea. While beneficial phytochemicals to human have revealed in cereals, the information on peptides from cereals is far less accumulated than major reserve protein. Here, we analyzed peptide profiles using surface enhanced laser desorption/ionization time of flight mass spectrometry (SELDI-TOF MS) in cereal seeds for construction of peptide information and attempted to develop peptide biomarkers for cereal identification. To optimize the analysis condition of SELDI-TOF MS, the effect of dilution factor on binding affinity to protein chips was tested using CM10 and Q10 arrays. Peptide clusters were significantly different at the level of 0.01 p-value. Peak spectra were the most stable in 1:50 of dilution factor in both chip arrays. Numbers of detected peak of 5 cereal seeds were 131 in CM10 and 74 in Q10 array. Each cereal was grouped as a cluster and well discriminated into different cluster in the level of 0.01 p-value. Numbers of potentially identified peptide biomarkers are 11, 13, 9, 5 and 12 in sorghum, foxtail millet, hog millet, adlay and corn, respectively. This study demonstrates that each cereal seed have own distinguishable specific peptides although their function are not identified yet in this study. In addition, the proteomic profiling using SELDI-TOF MS techniques could be a useful and powerful tool to discover peptide biomarker for discrimination and assess crop species, especially under 20 kDa.