• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2018.10a
• /
• pp.201-202
• /
• 2018

In this paper, a dog breed image provided by Dataset of existing ImageNet and Oxford-IIIT Pet Image is combined with a dog breed image obtained through data mining on Internet and a random-label is added. this paper introduces to recognize 122 classes of dog breeds and 1 class that is not dog breeds. The recognition rate of dog breeds using both conventional DB and collection DB was improved 1.5% over Top-1 compared to recognition rate of dog breeds using only existing DB. The image recognition rate about non-dog image, was 93% recognition rate in case of 10000 random DBs.

• Journal of Broadcast Engineering
• /
• v.27 no.5
• /
• pp.794-807
• /
• 2022

Although registration of dogs is mandatory according to the revision of the Animal Protection Act, the registration rate is low due to the inconvenience of the current registration method. In this paper, a performance improvement study was conducted on the dog face recognition technology, which is being reviewed as a new registration method. Through deep learning learning, an embedding vector for facial recognition of a dog was created and a method for identifying each dog individual was experimented. We built a dog image dataset for deep learning learning and experimented with InceptionNet and ResNet-50 as backbone networks. It was learned by the triplet loss method, and the experiments were divided into face verification and face recognition. In the ResNet-50-based model, it was possible to obtain the best facial verification performance of 93.46%, and in the face recognition test, the highest performance of 91.44% was obtained in rank-5, respectively. The experimental methods and results presented in this paper can be used in various fields, such as checking whether a dog is registered or not, and checking an object at a dog access facility.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.16 no.12
• /
• pp.8853-8860
• /
• 2015

This study is to identify recognition of and preference for toy breeds between young and older generations by investigating their understating on pet dogs, difference in pet preference and reasons behind their choice of dog breeds, recognition and preference by toy breeds. A survey was conducted among 137 youths and 60 elderlies in medium/large cities and rural areas. Collected data was processed with ${\chi}^2$-test to see statistical significance. The result showed a significance of p<0.01 in recognition of pet/companion animals and toy breeds, and in pet dog preference and reasons of dog choice, along with p<0.05 in recognition and preference by toy breeds. Thus, this research to provide basic information requires to widely understand characteristics of toy breeds throughout generations due to widespread preference for such breeds and to continue research on change factors in their level of recognition and preference for the right choice when he/she decides to raise a toy dog.

• The Korean Journal of Food And Nutrition
• /
• v.13 no.4
• /
• pp.365-371
• /
• 2000

After making a survey of edibility of dog meat on 963 male adults and 539 female adults, totalled 1,502 persons, the results were primarily divided into ages and sexes, dealing with statistics by Statistical Analys System. As a result, those who had eaten dog meat is average 83%, among them male adults is 91.9%, female adults is 67.9 %. The reason they have dog meat is as follows : in case of male adults "Following others going to dog meat restaurant and eating" shows high of 34.7 %, in female, "following family members who eat dog meat at home and eating" is most in the figure of 25.6%. Average 86.3 % favors edibility of dog meat, and among them, male is 92.3% and female 72.1%. The highest reason that male or female who opposed to edibility of dog meat is for being inhumane. 79.5% of male and 64.9 % of female know the fact that edible dog is exceptionally bred. The respondents answered most that dog meat was purchased from market. The first reason for objection to the criticism of eating dog meat is that male and female commonly answered most that ′As dog meat food Is our traditional food culture. it is not the problem to be found fault with by others.′ The second reason for that is followed by iris deliciousness.

• The Korean Journal of Food And Nutrition
• /
• v.13 no.4
• /
• pp.372-378
• /
• 2000

After making survey of edibility of dog meat on 963 male adults and 539 female adults, totalled 1,502 persons, the results were primarily divided into ages and sexes. dealing wish statistics by Statistical Analysis System. As a result, dog meat food most favored is Bosintang(dog meat soup), followed by Jeongol(dog meat stew), Suyuk(boiled dog meat), Muchim(boils dog meat added by spice and mixed). The frequency of having dog meat is two or three times a year. The age of having firstly had dog meat is most at the age of 21∼30 in male, and in female, 11∼20. Among dog meat cookery of Chosun dynasty known by respondents, Gaejang(dog meat soup) is most, and Musulzu(wine made from dog meat), Ott-bosintang(dog meat soup boiled with lacker tree), Pyeonyuk(boiled and sliced dog meat) follow respectively. It shows that the largest number of respondents answered what was improved after having dog meat was to \`become healthy, \`followed by \`become energetic\`. It reveals that dog meat cuisine desired to be newly developed was roasted dog meat, on which respondents answered most, followed by Tangsuyuk (fried dog meat served with syrup) and impromptu Bosintang.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.2
• /
• pp.99-106
• /
• 2022

In this paper, We propose a method for identifying objects by setting inscriptions and landmarks of dogs. The phenomenon of abandoning dogs is on the rise, and the number of abandoned individuals is also rapidly increasing. These abandoned dogs are becoming wild animals, causing a lot of damage to people's daily life, causing serious problems. As a solution to this problem, the animal registration system is being implemented, but there is a phenomenon that some dog owners avoid the registration method that inserts a chip, so the complete registration system is not settled. When registering a dog, removing the avoidance of dog owners will help establish the companion animal registration system. In this paper, we present a technique to identify objects by setting inscriptions and landmarks of dogs so that dog owners can register their dogs in a friendly way to eliminate this avoidance phenomenon.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.12
• /
• pp.1-9
• /
• 2020

As pet dogs rapidly increase in number, abandoned and lost dogs are also increasing in number. In Korea, animal registration has been in force since 2014, but the registration rate is not high owing to safety and effectiveness issues. Biometrics is attracting attention as an alternative. In order to increase the recognition rate from biometrics, it is necessary to collect biometric images in the same form as much as possible-from the face. This paper proposes a method to determine whether a dog is facing front or not in a real-time video. The proposed method detects the dog's eyes and nose using deep learning, and extracts five types of directional face information through the relative size and position of the detected face. Then, a machine learning classifier determines whether the dog is facing front or not. We used 2,000 dog images for learning, verification, and testing. YOLOv3 and YOLOv4 were used to detect the eyes and nose, and Multi-layer Perceptron (MLP), Random Forest (RF), and the Support Vector Machine (SVM) were used as classifiers. When YOLOv4 and the RF classifier were used with all five types of the proposed face orientation information, the face recognition rate was best, at 95.25%, and we found that real-time processing is possible.

• Journal of Legislation Research
• /
• no.54
• /
• pp.261-301
• /
• 2018

In South Korea, the number of households that rear companion animal has been gradually increasing. With the rise of household dogs, the frequency of humans that are being bitten by another's dog has also increased. This type of accident, which is known as a dog bite accident throughout the United States. It can cause significant physical and emotional damage to the victims and may result in grave injuries or death. Dog bite accidents are serious public health problems and can cause immeasurable hidden costs to the community. South Korea has enacted several laws to address dog bite accidents, which include the Animal Protection Act, the Civil Act, and the Criminal Act. On March 20, 2018, the Animal Protection Act was amended to reinforce the current legislation. These amendments addressed the duty of care owed by a companion dog owner to society members and the punishment that an owner of a fierce dog would face in the event of a dog bite accident. Conversely, several states in the United States have enacted a single law that regulates the details regarding dog bite accidents, such as the type of dog or animal, the type of damage, the scope of compensation for damages, and the scope of recognition of liabilities. This paper is intended to review the present situation of dog bite legislation in several states in the United States, which have a variety of laws that address dog bite accidents, and compare them with current South Korean dog bite legislation. Through this research, this paper will discuss what issues may exist in South Korean's current dog bite laws, analyze the responsibility of companion dog owners, and provide solutions to any issues that are discovered.

• Proceedings of the Korean Society of Veterinary Clinics Conference
• /
• 2008.10a
• /
• pp.175-175
• /
• 2008

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2023.07a
• /
• pp.519-520
• /
• 2023

본 논문에서는 최근 반려동물 등록제가 확대되고 있는 바, 기존의 마이크로 칩 삽입 방법을 회피하고 반려견 이미지를 통하여 개체를 인식하는 방법을 연구하였다. 반려견의 전체 이미지를 학습시켜 해당 개체를 식별하는 지능형 시스템을 ResNet 알고리즘을 이용하여 구현하고, 수집된 반려견의 개체 사진을 학습시켜 필요한 개체를 식별할 수 있도록 하였다.