Research on Similar Animal Classification Based on CNN Algorithm

Animal image classification with CNN (convolutional neural network) is commonly investigated in aera of image recogniation and classification, but major studies focus on species pictures classification with obvious distinctions. For example, CNN is usually employed to distinghish images between dogs and cats. This article puts the effort on similar animal images classification by applying simple 2D CNN via python. It focus on the binary classification for snub-nosed monkeys and normal monkeys. This distinguishment is hard to be done manually in a short time. For constructing complete convolutional neural network, some preparations are done in advance, such as the database construction and preprocess. The database is constructed by python crawler (downloading from google images), with 800 and 200 images for each class respectively as train data and test data. The pre-work includes image resizing, decoding and standardization. After that, the model is trained and then tested for verifying the model reliability. The training accuracy is 96.67% without any abnormality. On the basis of successful training, the test accuracy almost coincides with train accuracy in each 50 generations and plots in a graph. It indicates similar trends and results for them in the whole process. Because of this, CNN model in the study can help people identify rare animals in time and then people can effectively protect them. Therefore, CNN will be helpful in field of animal conservation, especially for rare species.

Zoomorphic Images of the BII Stage Trypillia Settlements in the Middle Buh River Region

Widely known in Trypillia archaeological culture various zoomorphic images clearly represent multisided spiritual world of farmers of Eneolithic period, in particular the spread of domestic animals cult. While retaining common features of zoomorphism throughout the whole range of Trypillia, from the early to later stages, there are certain distinctive features for each region and time of existence. Their study allows us to understand the peculiarities of formation and development of the animal image, manners and styles of its performance. It is important to study the embodiments of the animal image in various types of archaeological materials of Trypillia settlements of the Middle Buh River region at BII stage (plastics, pottery, dwelling models) and determine the territorial and temporal features in comparison with other areas of the Trypillia spread. At the settlements of Voroshylivka, Sosny, Selyshche, Nemyriv and Karolina, the most found plastic items are statuettes of domestic animals, which were made in realistic or stylistic manners. Small and medium-sized figurines, which are classified as bovidae/ovis, dominate here. They had paired legs, small horns and a tail. Sometimes an udder or a penis was imitated, indicating the sex of the animal (cow, bull). Of particular interest are the statuettes and zoomorphic bowls, which had cross holes in their coupled legs. It is possible that they were supplemented with wheels and used in rituals as movable figurines. Common are zoomorphic bowls on four legs with a horned head and a tail on rims, as well as peculiar oval or rectangular bowls on four legs, reminiscent of cattle feeding troughs. They were painted in a zoomorphic ornament, resembling the spine or fur of an animal. It is important to note that no “dog”, “deer” or “bird” ornaments, known in other regions of the Trypillia culture, have been found on the tableware of the Middle Buh local group. There are also zoomorphic features on the cooking ware. The pots had four legs on the bottoms, and small horned heads were stuck on the rims. The zoomorphism can be traced on a well-known dwelling model from Voroshylivka, where the roof timbers above the front and back walls are designed like the horns of a domestic animal. Considering that the model had four high legs, it is reasonable to assume that it could have been used in rituals honoring the dwelling and livestock. The comprehensive analysis of zoomorphic images from the settlements of the Trypillia BII stage of the Middle Buh local group showed some regional differences as well as general trends in the development of ideological views of the ancient agricultural population of the Right-Bank Ukraine. The image of animals, especially cattle, deeply penetrated into consciousness of the holders of Trypillia culture, and the cult of the horned bull became one of the leading in their ideology.

Study on the Taste-oriented Design of Ceramic Doll in Foshan City

Foshan city of Guangdong province, who enjoys the reputation of Ceramics’ Capital in South China is one of the largest and most important ceramic production bases in China. Foshan ceramic works of art have a long history and have a great influence in the whole country. In this paper, we try to change the design of the simple historical figure, zodiac animal image and vase purely for daily necessities. To enrich the theoretical results of Foshan ceramic doll art research, Foshan ceramic doll insists taste-oriented design so as to further strengthen the research of interesting design, better adapt to the psychological needs of the audience, create the popular literary and artistic works, lead to the improvement of people’s consuming taste, and finally promote the construction of social spiritual civilization and the development of cultural industry in Foshan.

Identification of Molluscan Animal Image in Two-Dimensional Print Media

Humans have known mollusks for a long time. The diverse and unique shell shapes are interesting to draw. The easiest medium to describe the shape of a mollusk is in two dimensions. This study aims to identify various images of mollusks in two-dimensional print media such as cloth, paper and plates. Based on the 10 sources of photos analyzed, 56 species of mollusks from 38 families were identified. The Gastropod class dominates with 45 species from 31 families, followed by Bivalves with 7 species from 5 families, then Cephalopods with 4 species from 2 families. Some of the problems found are the shape and proportion of images that different with specimens, some inverted or cropped images, different direction of rotation of the shells with specimens, and different colour patterns with specimens. Biological and distributional aspects of several families will be discussed briefly in this paper.

PENERAPAN PENDEKATAN KONTEKSTUAL MENGGUNAKAN MEDIA GAMBAR UNTUK MENINGKATKAN KETERAMPILAN MENULIS DESKRIPSI PADA SISWA KELAS II SEKOLAH DASAR

This research was conducted because it was found that there were problems in the Indonesian language subject for grade II students of Bhoanawa 2 Impres Elementary School which showed very low Indonesian learning outcomes. The formulations of the problems in this study are: 1) Does learning through a contextual approach by utilizing animal image media can improve writing writing skills ?; 2) Can a contextual approach using animal pictures media increase student activity? The data collection methods used were: observation, interview, test and documentation. The subjects in this study were all students in grade II of the Bhoanawa 2 Impres Elementary School, totaling 17 people. This type of research is Classroom Action Research. The results showed that the increase in writing skills in writing was 29%, or only 5 out of 17 students who were able to write well, which could achieve a greater value of 67. The results of cycle I increased to 47.05% (less) because of 17 students 8 students who managed to get a score above the value of the Minimum Completeness Criteria, and in cycle II an increase of 88.23% (very good), of the 17 students sampled there were 5 students in the very good category and 10 students in the good category.

Ontology-guided Semantic Composition for Zero-shot Learning

Zero-shot learning (ZSL) is a popular research problem that aims at predicting for those classes that have never appeared in the training stage by utilizing the inter-class relationship with some side information. In this study, we propose to model the compositional and expressive semantics of class labels by an OWL (Web Ontology Language) ontology, and further develop a new ZSL framework with ontology embedding. The effectiveness has been verified by some primary experiments on animal image classification and visual question answering.