Abstract: This paper presents a method for improving object search accuracy using a deep learning model. A major limitation to provide accurate similarity with deep learning is the requirement of huge amount of data for training pairwise similarity scores (metrics), which is impractical to collect. Thus, similarity scores are usually trained with a relatively small dataset, which comes from a different domain, causing limited accuracy on measuring similarity. For this reason, this paper proposes a deep learning model that can be trained with a significantly small amount of data, a clustered data which of each cluster contains a set of visually similar images. In order to measure similarity distance with the proposed method, visual features of two images are extracted from intermediate layers of a convolutional neural network with various pooling methods, and the network is trained with pairwise similarity scores which is defined zero for images in identical cluster. The proposed method outperforms the state-of-the-art object similarity scoring techniques on evaluation for finding exact items. The proposed method achieves 86.5% of accuracy compared to the accuracy of the state-of-the-art technique, which is 59.9%. That is, an exact item can be found among four retrieved images with an accuracy of 86.5%, and the rest can possibly be similar products more than the accuracy. Therefore, the proposed method can greatly reduce the amount of training data with an order of magnitude as well as providing a reliable similarity metric.
Abstract: In this paper, we are interested in the problem of
finding similar images in a large database. For this purpose we
propose a new algorithm based on a combination of the 2-D
histogram intersection in the HSV space and statistical moments. The
proposed histogram is based on a 3x3 window and not only on the
intensity of the pixel. This approach overcome the drawback of the
conventional 1-D histogram which is ignoring the spatial distribution
of pixels in the image, while the statistical moments are used to
escape the effects of the discretisation of the color space which is
intrinsic to the use of histograms. We compare the performance of
our new algorithm to various methods of the state of the art and we
show that it has several advantages. It is fast, consumes little memory
and requires no learning. To validate our results, we apply this
algorithm to search for similar images in different image databases.
Abstract: Functioning of a biometric system in large part
depends on the performance of the similarity measure function.
Frequently a generalized similarity distance measure function such as
Euclidian distance or Mahalanobis distance is applied to the task of
matching biometric feature vectors. However, often accuracy of a
biometric system can be greatly improved by designing a customized
matching algorithm optimized for a particular biometric application.
In this paper we propose a tailored similarity measure function for
behavioral biometric systems based on the expert knowledge of the
feature level data in the domain. We compare performance of a
proposed matching algorithm to that of other well known similarity
distance functions and demonstrate its superiority with respect to the
chosen domain.