http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
A Robust Face Detection Method Based on Skin Color and Edges
Ghimire, Deepak,Lee, Joonwhoan Korea Information Processing Society 2013 Journal of information processing systems Vol.9 No.1
In this paper we propose a method to detect human faces in color images. Many existing systems use a window-based classifier that scans the entire image for the presence of the human face and such systems suffers from scale variation, pose variation, illumination changes, etc. Here, we propose a lighting insensitive face detection method based upon the edge and skin tone information of the input color image. First, image enhancement is performed, especially if the image is acquired from an unconstrained illumination condition. Next, skin segmentation in YCbCr and RGB space is conducted. The result of skin segmentation is refined using the skin tone percentage index method. The edges of the input image are combined with the skin tone image to separate all non-face regions from candidate faces. Candidate verification using primitive shape features of the face is applied to decide which of the candidate regions corresponds to a face. The advantage of the proposed method is that it can detect faces that are of different sizes, in different poses, and that are making different expressions under unconstrained illumination conditions.
Extreme Learning Machine Ensemble Using Bagging for Facial Expression Recognition
Ghimire, Deepak,Lee, Joonwhoan Korea Information Processing Society 2014 Journal of information processing systems Vol.10 No.3
An extreme learning machine (ELM) is a recently proposed learning algorithm for a single-layer feed forward neural network. In this paper we studied the ensemble of ELM by using a bagging algorithm for facial expression recognition (FER). Facial expression analysis is widely used in the behavior interpretation of emotions, for cognitive science, and social interactions. This paper presents a method for FER based on the histogram of orientation gradient (HOG) features using an ELM ensemble. First, the HOG features were extracted from the face image by dividing it into a number of small cells. A bagging algorithm was then used to construct many different bags of training data and each of them was trained by using separate ELMs. To recognize the expression of the input face image, HOG features were fed to each trained ELM and the results were combined by using a majority voting scheme. The ELM ensemble using bagging improves the generalized capability of the network significantly. The two available datasets (JAFFE and CK+) of facial expressions were used to evaluate the performance of the proposed classification system. Even the performance of individual ELM was smaller and the ELM ensemble using a bagging algorithm improved the recognition performance significantly.
Security Verification of Video Telephony System Implemented on the DM6446 DaVinci Processor
Ghimire, Deepak,Kim, Joon-Cheol,Lee, Joon-Whoan The Korea Contents Association 2012 International Journal of Contents Vol.8 No.1
In this paper we propose a method for verifying video in a video telephony system implemented in DM6446 DaVinci Processor. Each frame is categorized either error free frame or error frame depending on the predefined criteria. Human face is chosen as a basic means for authenticating the video frame. Skin color based algorithm is implemented for detecting the face in the video frame. The video frame is classified as error free frame if there is single face object with clear view of facial features (eyes, nose, mouth etc.) and the background of the image frame is not different then the predefined background, otherwise it will be classified as error frame. We also implemented the image histogram based NCC (Normalized Cross Correlation) comparison for video verification to speed up the system. The experimental result shows that the system is able to classify frames with 90.83% of accuracy.
Data-Driven Intelligent Feeding System for Pet Care
Ghimire Ravi,Jae Weon Choi 제어로봇시스템학회 2022 제어로봇시스템학회 국제학술대회 논문집 Vol.2022 No.11
The rapid development of artificial intelligence, the internet of things, and digital information processing technology has a huge impact on our daily lives with smart devices and wearables. The well-being of companion animals such as dogs and cats has become a large challenge. An increasing number of pet owners, their emotional attachment with their pets, and the 21st-century’s lifestyle importantly need the safety and welfare of pets by harnessing a smart technological approach. This paper analyzes and compares different machine learning algorithms for data-driven intelligent feeding system for pet care application. Different parameters such as body weight growth, body temperature, heart rate, eating habits, activity, sleep, and urine pH have been considered with other correlated sub-variables in creating virtual datasets. The supervised machine learning models: linear regression, gaussian process regression, narrow neural network, linear support vector machine, and fine tree are evaluated and discussed for estimating feed quantity. The machine learning model was verified by training, validation, and testing datasets. The developed model will be an innovative breakthrough for pet care applications. Feed estimation can be automated using the pet’s health parameters, this will help the pet to prevent obesity and related disorders.
Ghimire, B.K.,Yoo, J.H.,Yu, C.Y.,Chung, I.M. Elsevier 2017 Asian Pacific journal of tropical medicine Vol.10 No.7
<P>Objective: To investigate the composition of volatile compounds in the different accessions of Perillafruteseens (P. frittescens) collected from various habitats of China and Japan. Methods: In the present study, the essential oil from the leaves of P. frutescens cultivars from China and Japan was extracted by hydro-distillation and the chemical composition and concentration of the volatile components present in the oils were determined by gas chromatography mass spectrometry (GC-MS) analysis. Results: Among the volatile components, the major proportion was of perilla ketone, which was followed by elemicin and beta-caryophyllene in the Chinese Perilla cultivars. The main component in the oil extracted from the Japanese accessions was myristicin, which was followed by perilla ketone and beta-caryophyllene. We could distinguish seven chemotypes, namely the perilla ketone (PK) type, perilla ketone, myristicin (PM) type, perilla ketone, unknown (PU) type, perilla ketone, beta-caryophyllene, myristicine (PB) type, perilla ketone, myristicin, unknown (PMU) type, perilla ketone, clemicine, myristicin, beta-caryophyllene (PEMB) type, and the perilla ketone, limonene, beta-cryophyllenc, myristicin (L) type. Most of the accessions possessed higher essential oil content before the flowering time than at the flowering stage. The average plant height, leaf length, leaf width of the Chinese accessions was higher than those of the Japanese accessions. Conclusion: The results revealed that the harvest time and geographical origin caused polymorphisms in the essential oil composition and morphological traits in the Perilla accessions originating from China and Japan. Therefore, these chemotypes with desirable characters might be useful for industrial exploitation and for determining the harvest time.</P>
Embryological studies on <i>Abelia tyaihyoni</i> Nakai (Caprifoliaceae)
Ghimire, Balkrishna,Suh, Gang Uk,Lee, Cheul Ho,Heo, Kweon,Jeong, Mi Jin Elsevier 2018 Flora Vol.242 No.-
<P><B>Abstract</B></P> <P> <I>Abelia tyaihyoni</I> is a Korean endemic species and also designated as an endangered taxon in the IUCN Red List. We present a comprehensive embryology of <I>A. tyaihyoni</I>, comparing it with previously available information on the Caprifoliaceae and related families, and identifying possible evolutionary trends. Overall, comparisons showed that <I>Abelia</I> is similar to the other genera of Caprifoliaceae in many embryological features, including the tetrasporangiate anther, the anther wall, dicotyledonous wall formation, fibrous endothecium, simultaneous cytokinesis, tetrahedral tetrads, and three-celled pollen grains. In addition, <I>Abelia</I> share anatropous, unitegmic, and tenuinucellate ovules, ephemeral antipodals, and a cellular-type endosperm with many caprifoliaceous genera. Some embryological features, such as the amoeboid tapetum with several nuclei forming a polyploid mass, bisporic <I>Allium</I>-type embryo sac, obturator, bilayered endocarp, and compressed exotesta, are here described for the first time in this genus. The results confirm that <I>A. tyaihyoni</I> shares some significant embryological features with other Caprifoliaceae. Although previous embryological studies on the Caprifoliaceae lack convincing evolutionary explanations, our comparative approach demonstrates some possible apomorphies in <I>Abelia</I> such as the presence of two sterile carpels, bisporic <I>Allium</I>-type embryo sac, single-seeded fruit, bilayered and sclerified endocarp, and compressed exotesta.</P> <P><B>Highlights</B></P> <P> <UL> <LI> <I>Abelia</I> is similar to the other Caprifoliaceae in many embryological features. </LI> <LI> Amoeboid tapetum with several nuclei forming polyploidy mass is new to the genus. </LI> <LI> Bisporic <I>Allium</I> type of embryo sac and obturator formation is also new to the genus. </LI> <LI> <I>Allium</I> embryo sac, bilayered endocarp & compressed exotesta are possible apomorphies for the family. </LI> </UL> </P>