Papers - THI THI ZIN
-
Consumer Technology Perspective of a Trinomial Random Walk Model Reviewed
Thi Thi Zin, Pyke Tin, H. Hama
2019 IEEE 8th Global Conference on Consumer Electronics, GCCE 2019 758 - 759 2019.10
Authorship：Lead author, Corresponding author Language：English Publishing type：Research paper (international conference proceedings) Publisher：2019 IEEE 8th Global Conference on Consumer Electronics, GCCE 2019
Recently, the consumer technology association has widened the horizons of the consumer electronic society to include a variety of technologies such as the Internet of Things, Artificial Intelligence, vertical farming and etc. Moreover, among many others, the consumer technology highlights the importance of organic dairy cows and they are well beings so that timely based energy balance and body condition scores of individual cows are needed to be thoroughly studied. In this paper, we shall introduce a Trinomial Random Walk Model as an illustration of consumer technology for analyzing changes in dairy cow body conditions from time to time. In doing so, we will use the 5-point scale system with increment 0.25. Some experimental simulation results are presented by varying the parameters involved.
-
Image Technology and Matrix-Geometric Method for Automatic Dairy Cow Body Condition Scoring Reviewed
Thi Thi Zin, Pyke Tin, Y. Horii, I. Kobayashi
International Journal of Biomedical Soft Computing and Human Sciences 24 ( 1 ) 2019.7
Language：English Publishing type：Research paper (scientific journal)
Dairy cow body condition scoring is a key and important in nearly every aspect of modern precision dairy farms for management of welfare and healthcare, milk yields and split meals (feeding system), aware heat and reproduction peak, calving in smooth and calf saving in proof so that all in all profitability in use. However, the majority of dairy farmers are not doing the body condition scoring on regular basis due to lack of automation so that it causes time-consuming and very subjective. Thus in this paper, an automatic and efficient dairy cow body scoring system is proposed by using Image Technology and Matrix-Geometric Method. By fusing image technol-ogy and matrix-geometric method as a hybrid can lead to a new and efficient technique of dairy cow body condition scoring system. In order to do so, firstly, the anatomical cow body points are extracted from the back views and top views of cow images. Then the geometric properties of the extracted anatomical points are transformed into a Markov Chain Matrix to determine the body condition scores. For confirmation of the validity of the proposed method, some experimental results are shown by using a public cow body condition database.
-
Complex Human–Object Interactions Analyzer Using a DCNN and SVM Hybrid Approach Reviewed
Cho Nilar Phyo, Thi Thi Zin and Pyke Tin
Applied Sciences (Switzerland) 9 ( 9 ) 2019.5
Language：English Publishing type：Research paper (scientific journal) Publisher：Applied Sciences (Switzerland)
Nowadays, with the emergence of sophisticated electronic devices, human daily activities are becoming more and more complex. On the other hand, research has begun on the use of reliable, cost-effective sensors, patient monitoring systems, and other systems that make daily life more comfortable for the elderly. Moreover, in the field of computer vision, human action recognition (HAR) has drawn much attention as a subject of research because of its potential for numerous cost-effective applications. Although much research has investigated the use of HAR, most has dealt with simple basic actions in a simplified environment; not much work has been done in more complex, real-world environments. Therefore, a need exists for a system that can recognize complex daily activities in a variety of realistic environments. In this paper, we propose a system for recognizing such activities, in which humans interact with various objects, taking into consideration object-oriented activity information, the use of deep convolutional neural networks, and a multi-class support vector machine (multi-class SVM). The experiments are performed on a publicly available cornell activity dataset: CAD-120 which is a dataset of human-object interactions featuring ten high-level daily activities. The outcome results show that the proposed system achieves an accuracy of 93.33%, which is higher than other state-of-the-art methods, and has great potential for applications recognizing complex daily activities.
DOI： 10.3390/app9091869
-
Deep Learning for Recognizing Human Activities Using Motions of Skeletal Joints Reviewed
Cho Nilar Phyo, Thi Thi Zin, Pyke Tin
IEEE Transactions on Consumer Electronics 65 ( 2 ) 243 - 252 2019.5
Language：English Publishing type：Research paper (scientific journal) Publisher：IEEE Transactions on Consumer Electronics
With advances in consumer electronics, demands have increased for greater granularity in differentiating and analyzing human daily activities. Moreover, the potential of machine learning, and especially deep learning, has become apparent as research proceeds in applications, such as monitoring the elderly, and surveillance for detection of suspicious people and objects left in public places. Although some techniques have been developed for human action recognition (HAR) using wearable sensors, these devices can place unnecessary mental and physical discomfort on people, especially children and the elderly. Therefore, research has focused on image-based HAR, placing it on the front line of developments in consumer electronics. This paper proposes an intelligent HAR system which can automatically recognize the human daily activities from depth sensors using human skeleton information, combining the techniques of image processing and deep learning. Moreover, due to low computational cost and high accuracy outcomes, an approach using skeleton information has proven very promising, and can be utilized without any restrictions on environments or domain structures. Therefore, this paper discusses the development of an effective skeleton information-based HAR which can be used as an embedded system. The experiments are performed using two famous public datasets of human daily activities. According to the experimental results, the proposed system outperforms other state-of-the-art methods on both datasets.
-
Multivariate Stochastic Analyzer for Dairy Cow Body Condition Scoring Reviewed
Thi Thi Zin, Kosuke Sumi, Pyke Tin
Digital Image & Signal Processing (DISP'19) 2019.4
Language：English Publishing type：Research paper (international conference proceedings)
In this paper, we introduce a conceptual multivariate stochastic analyzer for assessing body condition scores of individual dairy cows. Specifically, by using digital image technologies and statistical stochastic methods, dairy cow body condition scoring machine is to be established. In modern precision dairy farming, the body condition score (BCS) plays an important role as an indicator for measuring health and wealth of a dairy farm. Based on the BCS, today dairy farm manage systems are improved in in various aspects such as milk production, right time for artificial insemination, prediction calving time and so on. Traditionally, human experts perform visual examinations on the key areas of cow body parts such as hook, pin bones, tail head, short ribs, and backbone starches for scoring. However, well-trained human experts become less and dairy farm sizes are bigger as a consequence manual body condition scoring is almost impractical. Thus, in this paper we propose an image technology based stochastic analyzer for automatic scoring the BCS measures of dairy cows. In order to do so, the proposed analyzer first extracts some key anatomical points of a cow by using two-dimensional images taken from top views. Then, the system will derive some distance and angular features of the anatomical points and employs stochastic sampling techniques for refining the extracted features to produce parameters of multiple regressive prediction models and to assess the body condition scores of all dairy cows in the farm. Finally, to confirm the validity of proposed analyzer, we perform some experiments by a well-known benchmark dataset. The experimental results seem to be promising with an impact of high accuracy.
-
Incorporating Digital Imaging in Dairy Cow Anatomical Features Detection Reviewed
Thi Thi Zin, Cho Nilar Phyo, Pyke Tin
Digital Image & Signal Processing (DISP'19) 2019.4
Language：English Publishing type：Research paper (international conference proceedings)
In today precision dairy farming, the most commonly used technologies include wearable devices which must be attached to cows in one way or another in order to monitor cow’s behaviors. Since the wearable sensors placements may be broken or lost and can burden additional stress to the cows, it is necessary to consider an alternative and effective non-contact monitoring system. For this purpose, digital imaging technologies are suitable due to their capabilities of continuous operation and able to full automation. Thus, in this paper, we propose a digital imaging approach based on topological persistence concepts to precision dairy cow monitoring system focused on automated dairy cow anatomical feature detection. Anatomically these features define as hips, hooks, pin bones, tail-heads and rear regions of the cow body. These features will be utilized in the decision making process if and where a cow is present in an image or video frame. Once the system detects a cow in the image, the system automatically identifies an individual cow. The proposed cow anatomical feature detection and cow identification have the potentials in detecting cow body conditions, health conditions in time, milk production trends and predicting calving time and heat occurrence. Finally, by using videos taken in a real-life cow farm, the experimental results confirm the validity of proposed method.
-
An Algebraic Geometric Imaging Approach for Automatic Dairy Cow Body Condition Scoring System Reviewed
Thi Thi Zin, Pyke Tin, Ikuo Kobayashi and Yoichiro Horii
World Academy of Science, Engineering and Technology (ICPDFTA 2019) 2019.4
Language：English Publishing type：Research paper (conference, symposium, etc.)
Today dairy farm experts and farmers have well recognized the importance of dairy cow Body Condition Score (BCS) since these scores can be used to optimize milk production, managing feeding system and as an indicator for abnormality in health even can be utilized to manage for having healthy calving times and process. In tradition, BCS measures are done by animal experts or trained technicians based on visual observations focusing on pin bones, pin, thurl and hook area, tail heads shapes, hook angles and short and long ribs. Since the traditional technique is very manual and subjective, the results can lead to different scores as well as not cost effective. Thus this paper proposes an algebraic geometric imaging approach for an automatic dairy cow BCS system.
-
Dairy Cow Body Conditions Scoring System Based on Image Geometric Properties Reviewed
Thi Thi Zin, Pyke Tin, Ikuo Kobayashi
2019 IEEE 1st Global Conference on Life Sciences and Technologies (LifeTech 2019) 2019.3
Language：English Publishing type：Research paper (international conference proceedings)
In modernized precision dairy farming, the importance of dairy cow body condition scores is well recognized for making the healthy, wealthy and optimizing milk production. Although a burst amount of researches have been investigated the condition scoring problems from various aspects, not much satisfactory results have been come out yet. So, this paper will propose a geometric imaging approach for an automatic dairy cow body conditions scoring system. Specifically, some significant land marks or anatomical points are to be extracted from the top view image of a cow and their geometrical properties such as angles, length and area are investigated to estimate body condition scores. In doing so, the proposed method will employ techniques of polynomial regression, multiple regression, Markov Chain classification. Finally, some experimental results will be presented by using self-collected datasets and some well-known public datasets. The performance of preliminary results shows promising so that the approach of proposed method can lead to be applicable in real life environments.
-
A Study on Detection of Precursor Behaviors of Estrus in Cattle Using Video Camera Reviewed
Hiromitsu Hama, Tetsuya Hirata, Tsubasa Mizobuchi, Thi Thi Zin
2019 IEEE 1st Global Conference on Life Sciences and Technologies (LifeTech 2019) 2019.3
Language：English Publishing type：Research paper (international conference proceedings)
Development of an estrus detection system by non-contact and non-invasive methods to improve productivity is a strong desire from livestock farmers with aging society. As one of the elemental technologies, we will focus on detecting precursor behaviors of estrus in cattle using video camera. First, we converted from two-dimensional motion on video image to three dimensional one. Next, some features which are well-known as estrus precursor behaviors, were selected, for example, walking speed, trajectory and relative positional relationship of two cattle. Through experimental results, we could confirm the effectiveness of our proposed algorism. As the result, although it is a small case, it was able to detect without any false positive and false negative.
-
OCR Perspectives in Mobile Teaching and Learning for Early School Years in Basic Education Reviewed
Thi Thi Zin, Swe Zar Maw, Pyke Tin
2019 IEEE 1st Global Conference on Life Sciences and Technologies (LifeTech 2019) 2019.3
Language：English Publishing type：Research paper (international conference proceedings)
In these days, teaching and learning systems in schools with the use of mobile or portable devices such as tablets, e-readers, smartphones are becoming keen interests of educators as well as parents and teachers in worldwide. In this aspect, the early years of school children in basic education are the most challenging and important in developing effective and quality education for life. Since they are quite young and unable to dedicate time the need for easy to use and effective learning aids has become vital. Especially their writing skills are somehow needed to be improved with encouragements. In order to do so, the handwritten of characters and numerals performed by the children especially those living in less developing countries should be correctly recognized so that means and ways for remedies and solutions for improvements could be found. Thus the optical character recognition techniques for handwritten alphabets and numerals are moving into front especially the handwritten of children of early years in schools. In this paper, we introduce a Mobile Tutor - an effective and correct way of character segmentation and recognition of messy and unclear handwritten characters to help children learn and practice handwriting and early numeric operations such as addition and subtraction as well as help teachers monitor and review children’s progress. Some experiments are performed by providing tablets to the users and collecting handwritten characters from the users for recognition and analysis.
-
OCR perspectives in mobile teaching and learning for early school years in basic education Reviewed
Thi Thi Zin, Swe Zar Maw, Pyke Tin
2019 IEEE 1st Global Conference on Life Sciences and Technologies, LifeTech 2019 173 - 174 2019.3
Language：English Publishing type：Research paper (international conference proceedings) Publisher：2019 IEEE 1st Global Conference on Life Sciences and Technologies, LifeTech 2019
In these days, teaching and learning systems in schools with the use of mobile or portable devices such as tablets, e-readers, smartphones are becoming keen interests of educators as well as parents and teachers in worldwide. In this aspect, the early years of school children in basic education are the most challenging and important in developing effective and quality education for life. Since they are quite young and unable to dedicate time the need for easy to use and effective learning aids has become vital. Especially their writing skills are somehow needed to be improved with encouragements. In order to do so, the handwritten of characters and numerals performed by the children especially those living in less developing countries should be correctly recognized so that means and ways for remedies and solutions for improvements could be found. Thus the optical character recognition techniques for handwritten alphabets and numerals are moving into front especially the handwritten of children of early years in schools. In this paper, we introduce a Mobile Tutor-an effective and correct way of character segmentation and recognition of messy and unclear handwritten characters to help children learn and practice handwriting and early numeric operations such as addition and subtraction as well as help teachers monitor and review children's progress. Some experiments are performed by providing tablets to the users and collecting handwritten characters from the users for recognition and analysis.
-
Dairy Cow Body Conditions Scoring System Based on Image Geometric Properties Reviewed
Thi Thi Zin, Pyke Tin, I. Kobayashi
2019 IEEE 1st Global Conference on Life Sciences and Technologies, LifeTech 2019 171 - 172 2019.3
Language：English Publishing type：Research paper (international conference proceedings) Publisher：2019 IEEE 1st Global Conference on Life Sciences and Technologies, LifeTech 2019
In modernized precision dairy farming, the importance of dairy cow body condition scores is well recognized for making the healthy, wealthy and optimizing milk production. Although a burst amount of researches have been investigated the condition scoring problems from various aspects, not much satisfactory results have been come out yet. So, this paper will propose a geometric imaging approach for an automatic dairy cow body conditions scoring system. Specifically, some significant land marks or anatomical points are to be extracted from the top view image of a cow and their geometrical properties such as angles, length and area are investigated to estimate body condition scores. In doing so, the proposed method will employ techniques of polynomial regression, multiple regression, Markov Chain classification. Finally, some experimental results will be presented by using self-collected datasets and some well-known public datasets. The performance of preliminary results shows promising so that the approach of proposed method can lead to be applicable in real life environments.
-
A study on detection of precursor behaviors of estrus in cattle using video camera Reviewed
H. Hama, T. Hirata, T. Mizobuchi, Thi Thi Zin
2019 IEEE 1st Global Conference on Life Sciences and Technologies, LifeTech 2019 166 - 167 2019.3
Language：English Publishing type：Research paper (international conference proceedings) Publisher：2019 IEEE 1st Global Conference on Life Sciences and Technologies, LifeTech 2019
Development of an estrus detection system by non-contact and non-invasive methods to improve productivity is a strong desire from livestock farmers with aging society. As one of the elemental technologies, we will focus on detecting precursor behaviors of estrus in cattle using video camera. First, we converted from two-dimensional motion on video image to three dimensional one. Next, some features which are well-known as estrus precursor behaviors, were selected, for example, walking speed, trajectory and relative positional relationship of two cattle. Through experimental results, we could confirm the effectiveness of our proposed algorism. As the result, although it is a small case, it was able to detect without any false positive and false negative.
-
A Hybrid Rolling Skew Histogram-Neural Network Approach to Dairy Cow Identification System Reviewed
Cho Nilar Phyo, Thi Thi Zin, H. Hama, I. Kobayashi
International Conference Image and Vision Computing New Zealand 2018-November 2019.2
Language：English Publishing type：Research paper (international conference proceedings) Publisher：International Conference Image and Vision Computing New Zealand
In this paper, we propose a hybrid method in which rolling skew histogram and neural network techniques are fused to recognize patterns and identify cows in the milking rotary parlor of dairy farms. Individual cow identification is very important for managing the welfare and health care of an individual cow and developing the body condition scoring system. Although there has some sensor-based cows' identification system, those systems require to attach the sensor devices on each cow which is costly and burden on the cows. Since the proposed method applies a single video camera which is a non-contact device for the identification of many different cow patterns, the proposed system is low cost and no burdens on the cow. In particular, the operation of the system takes place while the cows are in the milking process in rotary milking parlor where the monitoring of individual cow is more effective than some other time and places. The identification process is based on the black and white pattern on the cow's body while moving on the rotary milking parlor. For the detecting and cropping of cows' body region is carried out by using rolling the skew histogram and used for the training process of the deep convolutional neural network. The trained network is employed for the identification of individual cow in the testing process. The experiments are performed on the self-collected cow video dataset which includes around 60 different cow's body patterns that have been taken at the large-scale farm in Oita Prefecture, Japan. The experimental results show that the proposed system is promising with the overall accuracy of 96.3 % and it is very effective and practical for the real-time cow identification system needed for establishing a modern precision dairy farming.
-
Framework of Cow Calving Monitoring System Using a Single Depth Camera Reviewed
K. Sumi, Thi Thi Zin, I. Kobayashi, Y. Horii
International Conference Image and Vision Computing New Zealand 2018-November 2019.2
Language：English Publishing type：Research paper (international conference proceedings) Publisher：International Conference Image and Vision Computing New Zealand
Calving difficulty is the primary cause of the problem of increasing death loss in cow-calf. It also profoundly effects on the economic impact of farmers and producers because of calves' death, injury to cows and high veterinary cost. To help the calving difficult of cows in time, long time continuous monitoring is required for deciding when and how to assist the calving process of cows. On the other hand, the continuous monitoring of cows' welfare become the major burden task for labors especially in a large farm where has a large number of cows. Therefore, the demands of sophisticated technology for automatic monitoring of cows' welfare is more and more increasing every year. In recent years, some researcher develops the automatics monitoring and predicting the cows' calving behavior by using the sensors devices such as temperature sensors and acceleration sensors. However, the sensors based system has various problems such as spalling and malfunction of the sensors and even can cause the burden on the cow because sensors are needed to put inside or on the body of the cow. To overcome those problems, in this paper, we propose an automatic detection of cows' calving behavior by using the depth camera (3D camera) along with image processing and computer vision technology and coordinate system transformation concept. The proposed system by using 3D camera can reduce the burden of both labors and cows.
-
A Study on Abnormal Behavior Detection of Infected Shrimp Reviewed
Takehiro Morimoto, Thi Thi Zin, Toshiaki Itami
2018 IEEE 7th Global Conference on Consumer Electronics, GCCE 2018 818 - 819 2018.12
Language：English Publishing type：Research paper (international conference proceedings) Publisher：2018 IEEE 7th Global Conference on Consumer Electronics, GCCE 2018
A method of detecting infected shrimp has been developed. Though shrimp production is thriving in Japan, disease causes much damage. Because the cause is a virus infection, medical treatment is not currently possible. The infection route includes factors such as predation by environmental organisms and water-borne infection. However, no specific countermeasures have been developed. Therefore, early detection of infected shrimp is necessary to prevent secondary infection. When shrimp are infected, they exhibit the following abnormal behaviors: 1) not eating, 2) appearing in shallow water, or 3) making sudden movements. The developed method of detecting infected shrimp involves the use of image processing to determine when the shrimp are not eating.
-
Medication and Meal Intake Monitoring using Human-Object Interaction Reviewed
Pann Thinzar Seint, Thi Thi Zin, Mitsuhiro Yokota
2018 IEEE 7th Global Conference on Consumer Electronics, GCCE 2018 145 - 146 2018.12
Language：English Publishing type：Research paper (international conference proceedings) Publisher：2018 IEEE 7th Global Conference on Consumer Electronics, GCCE 2018
Needs for end-of-life care are rising because of increasing age-related challenges. Among them, maintenance of nutrition and medication play an important role in healthcare of elderly people. In this situation, caregivers need to receive the daily record as a trending of usage which will lead to improvements in the quality of care. Our objective is to establish a video monitoring system for the act of taking medication and eating activity which is designed by using human-object interaction. For the evaluation of medication intake scenario, we propose the hierarchical classification system using hybrid PRNN-SVM model for action classification and activity interpretation. By the contribution of rulebased learning, our system also recognizes drinking/eating activity.
-
Automatic Postmortem Human Identification using Collarbone of X-ray and CT Scan Images Reviewed
Hanni Cho, Thi Thi Zin, N. Shinkawa, R. Nishii, H. Hama
2018 IEEE 7th Global Conference on Consumer Electronics, GCCE 2018 369 - 370 2018.12
Language：English Publishing type：Research paper (international conference proceedings) Publisher：2018 IEEE 7th Global Conference on Consumer Electronics, GCCE 2018
Nowadays, human identification both before and after death is becoming one of the most important issues in various aspects. However, it is not easy to identify human by doctors or forensic experts manually because it consumes much time, especially in large scale victims. Therefore, an automatic human identification system becomes a vital need. For this purpose, we develop a computerized human identification system based on the collarbone of chest X-ray and CT scan images to identify an unknown person after death by using image processing technology.
-
An Automatic Estimation of Dairy Cow Body Condition Score Using Analytic Geometric Image Features Reviewed
Thi Thi Zin, Pyke Tin, I. Kobayashi, Y. Horii
2018 IEEE 7th Global Conference on Consumer Electronics, GCCE 2018 190 - 191 2018.12
Language：English Publishing type：Research paper (international conference proceedings) Publisher：2018 IEEE 7th Global Conference on Consumer Electronics, GCCE 2018
In today modern precision dairy farms, among many dominant factors the Body Condition Score (BCS) has been considered a critical value to optimize milk production, analyzing health problems, insemination timing, and many others. Currently, the BCS is measured by human experts giving time- consuming and varying outcomes from one expert to another so that an automatic estimation system for body condition scoring is needed to be developed. Although there have been some researchers on the topics of the BCS by using image processing techniques, an efficient and satisfactory method has not been found yet. Therefore, in this paper, a new approach to an automatic estimation of dairy cow BCS using analytic geometry image features will be considered. Some experimental results are shown by using the BCS database.
-
A Study on Non-contact and Non-invasive Neonatal Jaundice Detection and Bilirubin Value Prediction Reviewed
Sojiro Kawano, Thi Thi Zin, Yuki Kodama
2018 IEEE 7th Global Conference on Consumer Electronics, GCCE 2018 204 - 205 2018.12
Language：English Publishing type：Research paper (international conference proceedings) Publisher：2018 IEEE 7th Global Conference on Consumer Electronics, GCCE 2018
Neonatal jaundice is a yellowish discoloration of skin and eyes that commonly occurs in newborn babies. It is a physiological phenomenon in neonates and it occurs due to the overproduction of bilirubin, reduction of bilirubin treatment function. The quick and accurate treatment is required for neonatal jaundice because it can lead to nuclear jaundice, cerebral palsy, intellectual disturbance and various sequelae. The investigation methods for examining neonatal jaundice include examination using jaundice meter and blood sampling. However, these methods require continuous monitoring and can cause burden on newborn babies. In this paper, we propose the non-contact and non-invasive detection method for neonatal jaundice using image processing and computer vision technology. The experiments are performed on the data collected by University Hospital, University of Miyazaki. According to the experiments, we confirmed about the usefulness of proposed method which can work effectively for infants.