Input Query Image Research Articles

Thai botanical herbs and its characteristics: Using artificial neural network

This paper proposed an explicit supervising technique with artificial neural network (ANN) which is associated with typical Content-based image retrieval (CBIR) systems. The main purpose of implementing this CBIR system with a unique supervising technique is for the system to apprehend various parts of Thai Botanical Herbs in the retrieving processing. This essential step prevents users from herbal usage misconception, and the lack of proper knowledge may lead to faulty exploitation of botanical herbs in Thailand. The input image or query image for our CBIR system is an image data of botanical herb features. The implementation of ANN supervision provides four kinds of training to the retrieval system which consists of 35, 50, 70, and 100% image section training of the original image. The weight, color histogram and edge pattern are set manually in order for these data to be supervised with back propagation technique. The retrieving results manifested a 91% precision and 50% recall with our implemented technique for 50% training. With all the assumptions and procedures, the implemented herbal CBIR system can facilitate several botanical herb users in terms of proper knowledge and misconception.

Querying non‐uniform image databases for biometrics‐related identification applications

PurposeThis research project focuses on developing techniques and technologies for automatically identifying human faces from images in the situations where face sample collections in the database as well as in the input query images are “as is”, i.e. no standard data collection environment is available. The developed method can also be used in other biometric applications.Design/methodology/approachThe specific method presented in this paper is called scale independent identification (SII). SII allows direct “comparison” between two images in terms of whether the two objects (e.g. faces) in the two images are the same object (i.e. the same individual). SII is developed by extensively using the matrix computation theory and in particular, the singular value decomposition theory.FindingsIt is found that almost all the existing methods in the literature or technologies in the market require that a normalization in scale be done before any identification processing. However, it is also found that normalization in scale not only adds additional processing complexity, but also may reduce the identification accuracy. In addition, it is difficult to anticipate an “optimal” scale in advance. The developed SII complements the existing methods in all these aspects.Research limitations/implicationsThe only limitation which is also the limitation for many other biometric identification methods is that each object (e.g. individual in human face identification) must have a sufficient number of training samples collected before the method works well.Practical implicationsSII is particularly suitable in law enforcement and/or intelligence applications in which it is difficult or impossible to collect data in a standard, “clean” environment.Originality/valueThe SII method is new, and the paper should be interesting to researchers or engineers in this area, and should also be interesting to companies developing any biometrics‐based identification technologies as well as government agencies.