scholarly journals Developing Forecasting Model in Thailand Fashion Market Based on Statistical Analysis and Content-Based Image Retrieval

2017 ◽  
pp. 159-173
Author(s):  
Komaek Kawinakrathiti ◽  
Suphakant Phimoltares ◽  
Patcha Utiswannakul

Traditional trend forecasting process in Thailand fashion industry was challenged by a fast fashion. In this paper, the Content-Based Image Retrieval (CBIR) technique is utilized for retrieval of a fashion trendsetter in fast fashion influence. Firstly, six fashion theories were implemented as 12 variables affecting the trendsetter. Cluster analysis, and factor analysis approach were used to find out the source of a fashion trendsetter as well. Cluster analysis separated all samples into three groups with different fashion ways. Moreover, factor analysis technique grouped all variables into three important factors. From such techniques, Internet media clearly is the best source of a fashion trendsetter. In the authors' model, traditional forecasting sources were added up with a fast fashion influence from CBIR. Then, the CBIR was evaluated in terms of efficiency compared with a real fashion expert in the Thai fashion industry. From statistical test, spatial color distribution yields high efficiency in selecting similar fashion style as a fashion expert.

2015 ◽  
Vol 5 (1) ◽  
pp. 32-46
Author(s):  
Komaek Kawinakrathiti ◽  
Suphakant Phimoltares ◽  
Patcha Utiswannakul

Traditional trend forecasting process in Thailand fashion industry was challenged by a fast fashion. In this paper, the Content-Based Image Retrieval (CBIR) technique is utilized for retrieval of a fashion trendsetter in fast fashion influence. Firstly, six fashion theories were implemented as 12 variables affecting the trendsetter. Cluster analysis, and factor analysis approach were used to find out the source of a fashion trendsetter as well. Cluster analysis separated all samples into three groups with different fashion ways. Moreover, factor analysis technique grouped all variables into three important factors. From such techniques, Internet media clearly is the best source of a fashion trendsetter. In the authors' model, traditional forecasting sources were added up with a fast fashion influence from CBIR. Then, the CBIR was evaluated in terms of efficiency compared with a real fashion expert in the Thai fashion industry. From statistical test, spatial color distribution yields high efficiency in selecting similar fashion style as a fashion expert.


Author(s):  
TIENWEI TSAI ◽  
YO-PING HUANG ◽  
TE-WEI CHIANG

In this paper, a two-stage content-based image retrieval (CBIR) approach is proposed to improve the retrieval performance. To develop a general retrieval scheme which is less dependent on domain-specific knowledge, the discrete cosine transform (DCT) is employed as a feature extraction method. In establishing the database, the DC coefficients of Y, U and V components are quantized such that the feature space is partitioned into a finite number of grids, each of which is mapped to a grid code (GC). When querying an image, at coarse classification stage, the grid-based classification (GBC) and the distance threshold pruning (DTP) serve as a filter to remove those candidates with widely distinct features. At the fine classification stage, only the remaining candidates need to be computed for the detailed similarity comparison. The experimental results show that both high efficacy and high efficiency can be achieved simultaneously using the proposed two-stage approach.


Author(s):  
T. Venkat Narayana Rao ◽  
Somasani Jyothi

In the recent era, as technology is growing rapidly the usage of social media is also increasing as a result large databases are required for storing the images. With the advancements in the technology, the storage of these images in computers has become possible. But retrieving the images is becoming a big task. We need to store them in a sequential manner and retrieve them when required. This paper details retrieval of images by considering the features related to content like shape, color, texture is called CBIR (content based image retrieval). As it is very difficult to extract the pictures in such huge data bases so we chose this technique which aim at high efficiency.


Author(s):  
Ramya Nemani

Cluster analysis is a mathematical technique in Multivariate Data Analysis which indicates the proper guidelines in grouping the data into clusters.  We can understand the concept with illustrated notations of cluster Analysis and various Clustering Techniques in this Research paper.  Similarity and Dissimilarity measures and Dendogram Analysis will be computed as required measures for Analysis.  Factor analysis technique is useful for understanding the underlying hidden factors for the correlations among the variables.  Identification and isolation of such facts is sometimes important in several statistical methods in various fields. We can understand the importance of the Factor Analysis and major concept with illustrated Factor Analysis approaches.  We can estimated the Basic Factor Modeling and Factor Loadings, and also Factor Rotation process.  Provides the complete application process and approaches of Principal Factor M.L.Factor and PCA comparison of Factor Analysis in this Research paper


2016 ◽  
Vol 16 (1) ◽  
pp. 85 ◽  
Author(s):  
I Gusti Rai Agung Sugiartha ◽  
Made Sudarma ◽  
I Made Oka Widyantara

Picture (image) is a media that used for storing visual data, for example, two-dimensional images are often used to store an incident. Images on the internet media growth very rapidly. There are a lot of image, video, text or other content on the Internet. Image Index and image retrieval again become a topic of research in the last decade in which concentrated on how to get the meaning of an information contained in an image. Three methods outlined in the search for an image, the text-based image retrieval, content-based image retrieval and indexing images in the order of language. This study focuses on the preparation of the features of an image based on color and texture. Features colors using the average value of Hue image, texture features using Gray Level occurance Matrix (GLCM). Color, texture, and shape extraction technique resulted in eighteen (18) feature that can be used as features in the process of Clustering.DOI: 10.24843/MITE.1601.12


2018 ◽  
Vol 9 (8) ◽  
pp. 660-665
Author(s):  
Chi Sheh ◽  
◽  
Peng Chan ◽  
Wen Jun Sim ◽  
◽  
...  

Fast fashion is becoming more and more popular nowadays and this industry is growing rapidly. In order to supply to the big demand of fast fashion clothing, company will need to increase the production of the clothing in shorter time frame. Besides that, to out beat the competitor, company will provide more choices of clothing in cheaper price to the customers. By practicing these actions to increase the business profits, company is behaving unethical to the manufacturer of the cloth. Most consumers are not aware of these ethical issues. This paper is will used and tested the conceptual model of fast fashion business ethics based on literature reviews. The finding from this paper will manifest the “real cost” of a cheap and branded fast fashion clothing and will be supported by real life event that happened. However, after realizing the problems, some company did make some changes and the solutions are stated in the paper as well.


2017 ◽  
Vol 5 (3) ◽  
pp. 54
Author(s):  
MOHAMMED ILIAS SHAIK ◽  
CHAUHAN DINESH ◽  
ESAPALLI SRINIVAS ◽  
PADIGE VINEETH ◽  
◽  
...  

Sign in / Sign up

Export Citation Format

Share Document