• Title/Summary/Keyword: handwriting

Search Result 178, Processing Time 0.029 seconds

Online Signature Verification using General Handwriting Data and CNN (일반 필기데이터와 CNN을 이용한 온라인 서명인식)

  • PARK, MINJU;YOUN, HEE YONG
    • Proceedings of the Korea Information Processing Society Conference
    • /
    • 2020.05a
    • /
    • pp.540-543
    • /
    • 2020
  • 본 논문에서는 대표적인 이미지 분류 모델인 CNN(Convolutional Neural Network)과 시간에 따른 이미지의 변화를 학습할 수 있는 LSTM(Long Short-Term Memory) 기반의 온라인 서명인식 모델을 제안한다. 실제로는 위조서명을 미리 구하기 어렵다는 사실을 고려해 서명검증 대상자가 아닌 타인의 진서명과 대상자의 일반 필기 데이터를 음의 데이터로서 학습에 사용하였다. 실험 결과, 전체 이미지 중 서명 부분의 비율에 따라 좋은 성능을 보이는 검증 모델이 다르며 Accuracy 성능지표를 통해 이 비율이 높거나 낮을 경우 CNN-LSTM 이, 중간일 경우 CNN 이 적합하다는 것을 확인하였다.

A Study on Hangeul Mobile Handwriting Practice and Analyzing Application Development Based on Deep Learning (딥러닝 기반 한글 전자 필기 연습 및 분석 앱 개발에 대한 연구)

  • Ko, Ju-Eun;Oh, Jee-Eun;Min, Kyoung-Won
    • Proceedings of the Korea Information Processing Society Conference
    • /
    • 2022.05a
    • /
    • pp.322-325
    • /
    • 2022
  • 전 세계적으로 코로나바이러스가 유행함에 따라 비대면 활동을 비롯하여 전자 필기 이용 및 상품 소비가 증가하였다. 전자 필기에 대한 수요가 늘어남에 따라 전자 필기 글씨체 교정에 대한 관심 또한 증가하는 추세이다. 본 논문에서는 전자 필기 이미지에서 음절과 음소 영역을 추출하여 글씨를 분석하고, 이를 사용하여 사용자의 손글씨에서 개선점을 찾아낼 수 있는 딥러닝 알고리즘을 제안한다. 제안한 알고리즘을 통해 사용자가 원하는 전자 필기 글씨체를 효과적으로 습득할 수 있도록 사용자 글씨에 대해 구체적인 피드백을 제공하는 딥러닝 기반 태블릿 PC 용 한글 전자 필기 연습 및 분석 앱에 대한 연구를 소개하였다.

On-line Handwriting Chinese Character Recognition for PDA Using a Unit Reconstruction Method (유닛 재구성 방법을 이용한 PDA용 온라인 필기체 한자 인식)

  • Chin, Won;Kim, Ki-Doo
    • Journal of the Institute of Electronics Engineers of Korea SP
    • /
    • v.39 no.1
    • /
    • pp.97-107
    • /
    • 2002
  • In this paper, we propose the realization of on-line handwritten Chinese character recognition for mobile personal digital assistants (PDA). We focus on the development of an algorithm having a high recognition performance under the restriction that PDA requires small memory storage and less computational complexity in comparison with PC. Therefore, we use index matching method having computational advantage for fast recognition and we suggest a unit reconstruction method to minimize the memory size to store the character models and to accomodate the various changes in stroke order and stroke number of each person in handwriting Chinese characters. We set up standard model consisting of 1800 characters using a set of pre-defined units. Input data are measured by similarity among candidate characters selected on the basis of stroke numbers and region features after preprocessing and feature extracting. We consider 1800 Chinese characters adopted in the middle and high school in Korea. We take character sets of five person, written in printed style, irrespective of stroke ordering and stroke numbers. As experimental results, we obtained an average recognition time of 0.16 second per character and the successful recognition rate of 94.3% with MIPS R4000 CPU in PDA.

A study on exposure of organic solvents in manufacturing industry (제조업 산업장의 유기용제 노출에 관한 연구)

  • Moon, Deog-Hwan;Kim, Jeong-Ha;Kim, Pil-Ja;Park, Myung-Hee;Hwang, Yong-sik;Lee, Chae-Kwan;Lee, Chang-Hee
    • Journal of Korean Society of Occupational and Environmental Hygiene
    • /
    • v.11 no.3
    • /
    • pp.219-228
    • /
    • 2001
  • Objective : For the purpose of preparing the fundamental data and health promotion and control program on organic solvents in air of manufacturing industry. Methods : The author surveyed number of organic solvent components which was used in working site and also determined the organic solvents concentration in air of 927 manufacturing industries and 1,267 working process with gas chromatography(NIOSH manual) for five years from 1995 to 1999. Results : Mean number of solvents components by type of industry, working process was number of 12. There were exceeded to TLV of 1,2-dichloroethane in textile manufacturing industry N,N-dimethyl furan in tanning and dressing of leather ; luggage, handbags, saddlery, harness and footwear manufacturing industry and chemical and chemical product manufacturing industry by type of industry. There were exceeded to TLV of 1,2-dichloroethane in handwriting and drawing process, cellosolve in adhesive spreading process, N,N-dimethly furan in production of solvent process and adhesion process by working process Total exceeded rate to threshold limit values of organic solvents mixture were 12.9% for EI(Exposure index) and 10.0% for Em(Estimation of mixture) by type of industry, 11.3% for EI and 8.2% for Em by type of working process. The highest exceed rate was 36.7% for EI in tanning and dressing of leather ; luggage handbags, saddlery, harness and footwear manufacturing industry and 29.0% for Em in textile manufacturing industry. The highest exceeded rate was 23.1% for EI and 12.5% for Em in adhesive spreading process by working process. Mean values of total subjects by type of industry and type of working process were $0.7{\pm}1.8$ for EI and $0.7{\pm}1.7$ for Em respectively. Conclusions : As above results, the author suggest that it makes the environmental control program on 1.2-dichloroethane, N,N-dimethyl furan, cellosolve by kind of organic solvent and on textile manufacturing industry, tanning and dressing of leather ; luggage, handbags, saddlery harness and footwear manufacturing industry by type of industry, and on handwriting, drawing process and adhesive spreading process and adhesion process by type of working process.

  • PDF

Neural Network-based Real-time End Point Detection Specialized for Accelerometer Signal (신경망을 이용한 실시간 가속도 신호 끝점 검출 방법)

  • Lim, Jong-Gwan;Kwon, Dong-Soo
    • 한국HCI학회:학술대회논문집
    • /
    • 2009.02a
    • /
    • pp.178-185
    • /
    • 2009
  • A signal processing algorithm is proposed for end point detection which is used commonly in accelerometers-based pattern recognition problem. In the conventional method, end points are detected by manual manipulation with an additive button or algorithm based on statistical computation and highpass filtering to cause critical time delay and difficulty for parameters optimization. As an solution, we propose a focused Time Lagged Feedforward Network(TLFN) with respect to a differential signal of acceleration, which is widely applied for time series forecasting. The simple experiment is conducted with handwriting and the detection performance and response characteristic of the proposed algorithm is tested and analyzed.

  • PDF

A Study on "Chijongbang(治瘇方)" (임언국(任彦國)의 "치종방(治瘇方)" 번역 연구)

  • Park, Sang-Young;Kwon, Oh-Min;Ahn, Sang-Young;Han, Chang-Hyun;Ahn, Sang-Woo
    • Journal of Korean Medical classics
    • /
    • v.22 no.4
    • /
    • pp.117-128
    • /
    • 2009
  • This study is purposed to list the full translation of "The secret method of tumor treatment(治瘇方)" and to report the results to the academic world. The results are described as follows. 1. It seems to be difficult to accept the current "The secret method of tumor treatment(治瘇秘方)" as the original script of "The method of tumor treatment(治瘇方)". It seemed that the Imeonguk(任彦國)'s 'Important and Valuable Method(遺方)' at the era of Joseon was copied by handwriting among the private sector. 2. There are contents which are overlapped with "Experience Method of Acupuncture(針灸經驗方)". A further study is required whether the "Experience Method of Acupuncture" was added later on. Because resolution for the before and after of these two books may be acted as an important matter to conjecture the related schools with acupuncture science. 3. The Korean style unique method and terminology were introduced frequently in "The method of tumor treatment". Besides that, materials for the medicine were filled with the things of easy-to-contact in our daily life. It is regarded as one of the excellent materials to confirm the superiority of oriental medical science through the clinical experiments today.

  • PDF

The Effects of Origami on the Improvement of Hand Dexterity

  • Bae, Ju Han
    • Journal of International Academy of Physical Therapy Research
    • /
    • v.4 no.2
    • /
    • pp.588-594
    • /
    • 2013
  • This study was carried out to investigate the effects of making an origami crane on the improvement of hand dexterity. Subjects composed of 20 normal adult males were randomly assigned to experimental and control groups of 10 people respectively. For the experimental group, a training of making an origami crane was conducted for 40 to 50 minutes a day during a 4-week training period. The control group was made to engage in everyday activities as usual. For pre and post assessment, Groove Pegboard test, Purdue Pegboard Test, and Jebsen Hand Function Test were used. The results on the effects of making an origami crane showed that there was a statistically significant difference in both the Grooved Pegboard test and Purdue Pegboard test(p<.05). In the Jebsen hand function Test, a significant difference was found in handwriting and building pieces of chess(p<.05), but there was no statistically significant difference in comparison with the right hand during the average performance of picking up small stuffs. The activity of making an origami crane for normal adults was confirmed to be helpful to improve the hand dexterity. Accordingly, making an origami crane is suggested to be an effective way to improve the hand dexterity.

Feature Extraction via Sparse Difference Embedding (SDE)

  • Wan, Minghua;Lai, Zhihui
    • KSII Transactions on Internet and Information Systems (TIIS)
    • /
    • v.11 no.7
    • /
    • pp.3594-3607
    • /
    • 2017
  • The traditional feature extraction methods such as principal component analysis (PCA) cannot obtain the local structure of the samples, and locally linear embedding (LLE) cannot obtain the global structure of the samples. However, a common drawback of existing PCA and LLE algorithm is that they cannot deal well with the sparse problem of the samples. Therefore, by integrating the globality of PCA and the locality of LLE with a sparse constraint, we developed an improved and unsupervised difference algorithm called Sparse Difference Embedding (SDE), for dimensionality reduction of high-dimensional data in small sample size problems. Significantly differing from the existing PCA and LLE algorithms, SDE seeks to find a set of perfect projections that can not only impact the locality of intraclass and maximize the globality of interclass, but can also simultaneously use the Lasso regression to obtain a sparse transformation matrix. This characteristic makes SDE more intuitive and more powerful than PCA and LLE. At last, the proposed algorithm was estimated through experiments using the Yale and AR face image databases and the USPS handwriting digital databases. The experimental results show that SDE outperforms PCA LLE and UDP attributed to its sparse discriminating characteristics, which also indicates that the SDE is an effective method for face recognition.

A Case Report of a Delayed Vascular Chorea Patient Treated with Sukjihwangbogan-tang-gagam (소뇌경색의 후유증으로 발생한 지연성 혈관 무도병 증례 보고 : 협진 치험 1례)

  • Ok, Hyo-joon;Park, Yeong-hwa;Lim, Bo-ra;Kwon, Do-ick;Choi, Min-ki;Nam, Su-hyun
    • The Journal of Internal Korean Medicine
    • /
    • v.37 no.5
    • /
    • pp.741-749
    • /
    • 2016
  • Objective: To evaluate the effect of Sukjihwangbogan-tang-gagam (熟地黃補肝湯加減) in a patient with delayed vascular chorea. Methods: A vascular chorea patient was treated with acupuncture, moxibustion, and herbal medicine (Sukjihwangbogan-tang-gagam). Results: Improvements in the Unified Huntington’s Disease Rating Scale (UHDRS), UFMG Sydenham’s Chorea Rating Scale (USCRS) (especially the scale of Activity of Daily Living (ADL)) (dysarthria 2→0, tongue protrusion 3→1, chorea 3→1, handwriting 3→1), and a Visual Analogue Scale (VAS) (8→2) were observed after the Sukjihwangbogan-tang-gagam treatment. Conclusions: Sukjihwangbogan-tang-gagam may be an effective treatment for patients with delayed vascular chorea.

The effects of the class management with Adler's logical consequences on elementary children's problem behavior (아들러의 '논리적 결과'를 활용한 학급운영이 초등학생의 문제 행동 개선에 미치는 효과)

  • Nam, Eun-Mi
    • The Korean Journal of Elementary Counseling
    • /
    • v.7 no.1
    • /
    • pp.135-163
    • /
    • 2008
  • This study intends to confirm if it is effective in improving the problem behaviors by applying "Logical Consequence" suggested by Adler to the class management. The researcher who is in charge of the 4th grade class of elementary school in G city applied Adlerian logical consequence to her children, observed and examined the effectiveness of reducing the problem behaviors by the qualitative study method of in-depth interview from March to October. The problem behaviors treated in this study includes irrelevant remarks and gossip in school time, no preparation for taking lessons in time, no preparation of a textbook and a supply, no involvement in doing homework, scribbling and poor handwriting on a textbook, teasing a friend (abusive language, joking, violence), indoor running, no involvement in doing a task, being late and no arrangement of indoor shoes. In conclusion, this study indicated that the use of Logical Consequence was relatively effective in improving the problem behaviors and more effective in individual behaviors rather than group's behaviors. While the problem behaviors conducted in a class in the daily routine were effective at the point of the occurred problems, the problem behaviors occurred at the point of time related to a home or the finish of daily routine were ineffective.

  • PDF