http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
- 中文 을 입력하시려면 zhongwen을 입력하시고 space를누르시면됩니다.
- 北京 을 입력하시려면 beijing을 입력하시고 space를 누르시면 됩니다.
RISS 인기검색어
- 검색키워드
- 저자 : 허찬 (검색결과 72 건)
- 무료
- 기관 내 무료
- 유료
-
-
허찬 연세대학교 국학연구원 2019 동방학지 Vol.187 No.-
This paper tried to look at the text of the typograhic versions of 「Sodaeseongjeon(蘇大成傳)」 published in the early days, and explore its formation process. In this paper, six types of typograhic version of 「Sodaeseongjeon」 published in the 1910s were analyzed and the features were explored. This process reconstructs the appearance process of typographic 「Sodaeseongjeon」 as follows; In 1914, two companies- Dongmiseosi(東美書市) in Keijo(京城) and Kwangmuncheaksa(光文冊肆) in Pyongyang(平壤) publishedtwo different typographic version of「Sodaeseongjeon」. Dongmiseosi has created a book that is easy to read by using the many abbreviated versions of Hanyang's woodblock version(京板坊刻本) text as its source, reducing lines and reducing embroidery. By comparison, Kwangmuncheaksa produced their version that was not readable because it had many lines or embroidery per face based on the original sequence, which was rich in content, not omitted or compressed, without missing pixels. Later, Keijo's companies published several 「Sodaeseongjeon」 in 1917, based on the Dongmiseosi script. In the case of the new forest, it was published by increasing the number of pages per page and reducing the ground while fully accepting the contents of the Dongmiseosi version. Kongjinseokwan(共進書館) and Shinkuseorim(新舊書林) attempted to differentiate themselves by transforming the story's haste based on the Dongmiseosi version, and similarly tried to increase profits by reducing the number of faces. In addition to transforming the haste, Pakmunseokwan(博文書館) inserted new characters and new pixels. 본 논문은 활판본 고소설이 출판되기 시작하였던 1910년대에 출현한 <소대성전> 이본을 살펴보고, 형성과정을 탐색하는 것을 목표로 하였다. <소대성전>은 현재 남아 있는 고소설 가운데 비교적 이른 시기에 출현한 중요한 작품으로, 지금까지 이본에 관한 유의미한 연구가 이루어졌다. 그러나 선행연구는 필사본과 방각본을 중심으로 이루어진 것으로, 20세기 이후의 유통과정에 관해 새롭게 규명될 필요가 있다. 이에 본 논문에서는 현재 남아 있는 1910년대 활판본 6종을 분석하고 특징을 찾아보고자 하였다. 이러한 과정을 통해 필자가 파악한 <소대성전> 활판본의 출현과정을 재구하면 다음과 같다. 1914년, 경성의 동미서시와 평양의 광문책사에서 활판본 <소대성전>을 출판하였다. 같은 제목의 작품을 간행한 두 출판사의 생각은 달랐던 것으로 보인다. 동미서시는 많은 축약이 이루어진 경판23장본계열의 이본을 원천으로 하여 한자를 붙이고, 한 면에 들어가는 분량을 줄여 독자들로 하여금 보기 편한 책을 만들어냈다. 이에 비해 광문책사에서는 내용이 풍부한 원본계열의 이본을 토대로 하여, 조판이 촘촘하게 이루어진, 그래서 면당 행수나 자수가 많은 <소대성전>을 만들어냈다. 이후 경성의 출판사들은 동미서시본을 토대로 하여 1917년에 몇 종의 <소대성전>을 간행하였다. 신구서림의 경우 동미서시본의 내용을 전적으로 수용하면서 면당 행수를 늘려 면수를 줄여 출판하였다. 신명서림과 공진서관은 동미서시본을 기반으로 이야기의 서두를 변형시켜 차별화를 시도했고, 마찬가지로 면수를 줄여 이윤을 제고하고자 하였다. 박문서관은 서두를 변형시키는 것에 더해, 새로운 인물과 새로운 화소를 추가하였다. 각각의 출판사들은 나름의 변화를 꾀하는 과정에서 완판이나 원본계열의 필사본 등, 경판23장본계열과 동미서시본 이외의 이본도 참고하였다. 경판23장본계열의 활판본 <소대성전>은 1920년대 이후에도 꾸준히 출판되었다. 그러나 광문책사본은 1917년 재판 이후로는 더 이상 유사한 텍스트를 생산하지 못하였다. 1914년에 나란히 출판된 두 활판본 <소대성전>은 전국적으로 유통되었던 활판본 고소설 시장에서 경쟁하였고, 승자는 동미서시본이라 할 수 있다. 이러한 현상이 갖는 의미에 대해서는 후속연구를 통해 보완하고자 한다.
-
허찬,김지현,김현욱,이주연,홍완수,김천제,백현동 한국축산식품학회 2010 한국축산식품학회지 Vol.30 No.1
The aim of this research was to estimate the effect of temperature and develop predictive models for the growth of total viable cells (TVC) and Escherichia coli (EC) on chicken breast under aerobic and various temperature conditions. The primary models were determined by Baranyi model. The secondary models for the specific growth rate (SGR) and lag time (LT), as a function of storage temperature, were developed by the polynomial model. The initial contamination level of chicken breasts was around 4.3 Log CFU/g of TVC and 1.0 Log CFU/g of E. coli. During 216 h of storage, SGR of TVC showed 0.05, 0.15, and 0.54 Log CFU/g/h at 5, 15, and 25℃. Also, the growth tendency of EC was similar to those of TVC. As storage temperature increased, the values of SGR of microorganisms increased dramatically and the values of LT decreased inversely. The predicted growth models with experimental data were evaluated by Bf, Af, RMSE, and R2. These values indicated that these developed models were reliable to express the growth of TVC and EC on chicken breasts. The temperature changes of distribution and showcase in markets might affect the growth of microorganisms and spoilage of chicken breast mainly.
-
Estimation of Shelf-life of Frankfurter Using Predictive Models of Spoilage Bacterial Growth
허찬,최윤상,Cheon-Jei Kim,백현동 한국축산식품학회 2009 한국축산식품학회지 Vol.29 No.3
The aim of this research was to develop predictive models for the growth of spoilage bacteria (total viable cells, Pseudomonas spp., and lactic acid bacteria) on frankfurters and to estimate the shelf-life of frankfurters under aerobic conditions at various storage temperatures (5, 15, and 25oC). The primary models were determined using the Baranyi model equation. The secondary models for maximum specific growth rate and lag time as functions of temperature were developed by the polynomial model equation. During 21 d of storage under various temperature conditions, lactic acid bacteria showed the longest lag time and the slowest growth rate among spoilage bacteria. The growth patterns of total viable cells and Pseudomonas spp. were similar each other. These data suggest that Pseudomonas spp. might be the dominant spoilage bacteria on frankfurters. As storage temperature increased, the growth rate of spoilage bacteria also increased and the lag time decreased. Furthermore, the shelf-life of frankfurters decreased from 7.0 to 4.3 and 1.9 (d) under increased temperature conditions. These results indicate that the most significant factor for spoilage bacteria growth is storage temperature. The values of Bf, Af, RMSE, and R2 indicate that these models were reliable for identifying the point of microbiological hazard for spoilage bacteria in frankfurters. The aim of this research was to develop predictive models for the growth of spoilage bacteria (total viable cells, Pseudomonas spp., and lactic acid bacteria) on frankfurters and to estimate the shelf-life of frankfurters under aerobic conditions at various storage temperatures (5, 15, and 25oC). The primary models were determined using the Baranyi model equation. The secondary models for maximum specific growth rate and lag time as functions of temperature were developed by the polynomial model equation. During 21 d of storage under various temperature conditions, lactic acid bacteria showed the longest lag time and the slowest growth rate among spoilage bacteria. The growth patterns of total viable cells and Pseudomonas spp. were similar each other. These data suggest that Pseudomonas spp. might be the dominant spoilage bacteria on frankfurters. As storage temperature increased, the growth rate of spoilage bacteria also increased and the lag time decreased. Furthermore, the shelf-life of frankfurters decreased from 7.0 to 4.3 and 1.9 (d) under increased temperature conditions. These results indicate that the most significant factor for spoilage bacteria growth is storage temperature. The values of Bf, Af, RMSE, and R2 indicate that these models were reliable for identifying the point of microbiological hazard for spoilage bacteria in frankfurters.
-
-
허찬,신석훈,이은복,지승도,온승엽 한국경영과학회 2015 한국경영과학회 학술대회논문집 Vol.2015 No.4
In this paper, we proposed a new method to search the optional set of wrinkle areas, which maximize the accuracy while reducing the number of areas. We calculate a wrinkle density of each of 8 wrinkle areas and use them to estimate the age of a face. Support Vector Machine is used to predict the age of a face using wrinkle densities as features for classification. Backward elimination method is used to reduce the number of features. We also provided experiment result, in which only 4 percent of accuracy is decreased while using only 2 features instead of 8.
-
-
생성 기반 캡션 정보가 결합된 비디오 특징을 이용한 텍스트-비디오 검색 성능 향상
허찬,이동훈,박혜영,박상효 한국정보과학회 2024 정보과학회 컴퓨팅의 실제 논문지 Vol.30 No.2
Text-to-video retrieval is an emerging research area that involves retrieving appropriate video clips in response to a text query. To achieve this purpose, researchers widely employ methods based on joint embedding, constructing a shared embedding space is constructed to represent the semantic meaning of both text and video data and utilizing them for retrieval. However, since the two heterogeneous data have inherently different characteristics, they will have different distributions in the joint embedding space, which may lead to degraded search performance. To overcome this problem, we propose a novel video feature representation method that combines visual and textual information. Specifically, the proposed model generates captions from the video and combines them with the visual features of the video to produce an enhanced video feature vector. These enhanced feature vectors improve the retrieval performance by reducing the modality gap between the query text and the candidate videos during the inference process. In experiments validating the efficiency of the proposed method, we observed a performance improvement of 3.7% (MSR-VTT) and 0.7% (VATEX) in the Recall@sum metric over the baseline model on two benchmark datasets.
-
-
연관 검색어 추천
이 검색어로 많이 본 자료
활용도 높은 자료
