Description:
In many real-world clustering problems, there usually exist little information about the clusters underlying a certain dataset. For example, the number of clusters hidden in many datasets is usually not known a priori. This is an issue because many traditional clustering methods require such information as input. This paper examines a practical stochastic clustering method (PSCM) that has the ability to find clusters in datasets without requiring users to specify the centroids or the number of clusters. By comparing with traditional methods (k-means, self-organising map and hierarchical clustering methods), the performance of PSCM is found to be robust against overlapping clusters and clusters with uneven sizes. The proposed method also scales well with datasets having varying number of clusters and dimensions. Finally, our experimental results on real-world data confirm that the proposed method performs competitively against the traditional clustering methods in terms of clustering accuracy and efficiency.
Description:
Finding clusters in data is a challenging problem. Given a dataset, we usually do not know the number of natural clusters hidden in the dataset. The problem is exacerbated when there is little or no additional information except the data itself. This paper proposes a general stochastic clustering method that is a simplification of nature-inspired ant-based clustering approach. It begins with a basic solution and then performs stochastic search to incrementally improve the solution until the underlying clusters emerge, resulting in automatic cluster discovery in datasets. This method differs from several recent methods in that it does not require users to input the number of clusters and it makes no explicit assumption about the underlying distribution of a dataset. Our experimental results show that the proposed method performs better than several existing methods in terms of clustering accuracy and efficiency in majority of the datasets used in this study. Our theoretical analysis shows that the proposed method has linear time and space complexities, and our empirical study shows that it can accurately and efficiently discover clusters in large datasets in which many existing methods fail to run.
Description:
Abstract— In image retrieval, an effective dissimilarity measure is required to retrieve the perceptually similar images. Minkowski-type (lp ) distance is widely used for image retrieval, however it has its limitations. It focuses on distance between image features and ignores the data distribution of the image features, which can play an important role in measuring perceptual similarity of images. !! also favours the most dominant components in calculating the total dissimilarity. A data dependent measure, named !! -dissimilarity, which estimates the dissimilarity using the data distribution, has been proposed recently. Rather than relying on geometric distance, it measures the dissimilarity between two instances in each dimension as a probability mass in a region that encloses the two instances. It considers two instances in a sparse region to be more similar than in a dense region. Using the probability of data mass enables all the dimensions of feature vectors to contribute in the final estimate of dissimilarity, so it does not just heavily bias towards the most dominant components. However, relying only on data distribution and completely ignoring the geometric distance raise another limitation. This can result in finding two instances similar only due to being in a sparse region, however if the geometric distance between them is large then they are not perceptually similar. To address this limitation we proposed a new hybrid data dependent dissimilarity (HDDD) measure that considers both data distribution as well as geometric distance. Our experimental results using Corel database and Caltech 101 show that (HDDD) leads to higher image retrieval performance than lp distance (lpD) and mp.
Description:
Abstract— In image retrieval, an effective dissimilarity measure is required to retrieve the perceptually similar images. Minkowski-type (lp ) distance is widely used for image retrieval, however it has its limitations. It focuses on distance between image features and ignores the data distribution of the image features, which can play an important role in measuring perceptual similarity of images. !! also favours the most dominant components in calculating the total dissimilarity. A data dependent measure, named !! -dissimilarity, which estimates the dissimilarity using the data distribution, has been proposed recently. Rather than relying on geometric distance, it measures the dissimilarity between two instances in each dimension as a probability mass in a region that encloses the two instances. It considers two instances in a sparse region to be more similar than in a dense region. Using the probability of data mass enables all the dimensions of feature vectors to contribute in the final estimate of dissimilarity, so it does not just heavily bias towards the most dominant components. However, relying only on data distribution and completely ignoring the geometric distance raise another limitation. This can result in finding two instances similar only due to being in a sparse region, however if the geometric distance between them is large then they are not perceptually similar. To address this limitation we proposed a new hybrid data dependent dissimilarity (HDDD) measure that considers both data distribution as well as geometric distance. Our experimental results using Corel database and Caltech 101 show that (HDDD) leads to higher image retrieval performance than lp distance (lpD) and mp.
Description:
In image retrieval, an effective dissimilarity (or similarity) measure is required to retrieve the perceptually similar images. Minkowski-type distance is widely used for image retrieval, however it has its limitation. It focuses on distance between image features and ignores the data distribution of the image features, which can play an important role in measuring perceptual similarity of images. To address this limitation, a data dependent measure named m-p, which calculates the dissimilarity using the data distribution rather than geometric distance has been proposed recently. It considers two instances in a sparse region to be more similar than in a dense region. Relying only on data distribution and completely ignoring the geometric distance raise other limitations. This may result in finding two perceptually dissimilar instances similar due to being located in a sparse region or vice versa. We proposed a new hybrid dissimilarity measure and experimental results show that it addresses these limitations.
Description:
Content-based image retrieval (CBIR) is a popular approach to retrieve images based on a query. In CBIR, retrieval is executed based on the properties of image contents (e.g. gradient, shape, color, texture) which are generally encoded into image descriptors. Among the various image descriptors, histogram-based descriptors are very popular. However, they suffer from the limitation of coarse quantization. In contrast, the use of kernel descriptors (KDES) is proven to be more effective than histogram-based descriptors in other applications, e.g. image classification. This is because, in the KDES framework, instead of the quantization of pixel attributes, each pixel equally takes part in the similarity measurement between two images. In this paper, we propose an approach for how the conventional KDES and its improved version can be used for CBIR. In addition, we have provided a detailed insight into the effectiveness of improved kernel descriptors. Finally, our experiment results will show that kernel descriptors are significantly more effective than histogram-based descriptors in CBIR.
Description:
A number of building detection methods have been proposed in the literature. However, they are not effective in detecting small buildings (typically, 50 m(2)) and buildings with transparent roof due to the way area thresholds and ground points are used. This paper proposes a new building mask to overcome these limitations and enables detection of buildings not only with transparent roof materials but also which are small in size. The proposed building detection method transforms the non-ground height information into an intensity image and then analyses the gradient information in the image. It uses a small area threshold of 1 m2 and, thereby, is able to detect small buildings such as garden sheds. The use of non-ground points allows analyses of the gradient on all types of roof materials and, thus, the method is also able to detect buildings with transparent roofs. Our experimental results show that the proposed method can successfully extract buildings even when their roofs are small and/or transparent, thereby, achieving relatively higher average completeness and quality.
Description:
A number of building detection methods have been proposed in the literature. However, they are not effective in detecting small buildings (typically, 50 m(2)) and buildings with transparent roof due to the way area thresholds and ground points are used. This paper proposes a new building mask to overcome these limitations and enables detection of buildings not only with transparent roof materials but also which are small in size. The proposed building detection method transforms the non-ground height information into an intensity image and then analyses the gradient information in the image. It uses a small area threshold of 1 m2 and, thereby, is able to detect small buildings such as garden sheds. The use of non-ground points allows analyses of the gradient on all types of roof materials and, thus, the method is also able to detect buildings with transparent roofs. Our experimental results show that the proposed method can successfully extract buildings even when their roofs are small and/or transparent, thereby, achieving relatively higher average completeness and quality.
Description:
Pairwise (dis) similarity measure of data objects is central to many applications of image anlaytics, such as image retrieval and classification. Geometric distance, particularly Euclidean distance ((
Description:
Pairwise (dis) similarity measure of data objects is central to many applications of image anlaytics, such as image retrieval and classification. Geometric distance, particularly Euclidean distance ((
Description:
Image representation using feature descriptors is crucial. A number of histogram-based descriptors are widely used for this purpose. However, histogram-based descriptors have certain limitations and kernel descriptors (KDES) are proven to overcome them. Moreover, the combination of more than one KDES performs better than an individual KDES. Conventionally, KDES fusion is performed by concatenating them after the gradient, colour and shape descriptors have been extracted. This approach has limitations in regard to the efficiency as well as the effectiveness. In this paper, we propose a novel approach to fuse different image features before the descriptor extraction, resulting in a compact descriptor which is efficient and effective. In addition, we have investigated the effect on the proposed descriptor when texture-based features are fused along with the conventionally used features. Our proposed descriptor is examined on two publicly available image databases and shown to provide outstanding performances.
Description:
Image representation using feature descriptors is crucial. A number of histogram-based descriptors are widely used for this purpose. However, histogram-based descriptors have certain limitations and kernel descriptors (KDES) are proven to overcome them. Moreover, the combination of more than one KDES performs better than an individual KDES. Conventionally, KDES fusion is performed by concatenating them after the gradient, colour and shape descriptors have been extracted. This approach has limitations in regard to the efficiency as well as the effectiveness. In this paper, we propose a novel approach to fuse different image features before the descriptor extraction, resulting in a compact descriptor which is efficient and effective. In addition, we have investigated the effect on the proposed descriptor when texture-based features are fused along with the conventionally used features. Our proposed descriptor is examined on two publicly available image databases and shown to provide outstanding performances.
Description:
This paper presents a novel method for registering multi-modal images, based on corners. The proposed method is motivated by the fact that large content differences are likely to occur in multi-modal images. Unlike traditional multi-modal image registration methods that utilize intensities or gradients for feature representation, we propose to use curvatures of corners. Moreover, a novel local descriptor called Distribution of Edge Pixels Along Contour (DEPAC) is proposed to represent the neighborhood of corners. Curvature and DEPAC similarities are combined in our method to improve the registration accuracy. Using a set of benchmark multi-modal images and multi-modal microscopic images, we demonstrate that our proposed method outperforms an existing state-of-the-art image registration method.
Description:
Similarity measure is an important research topic in image classification and retrieval. Given a type of image features, a good similarity measure should be able to retrieve similar images from the database while discard irrelevant images from the retrieval. Similarity measures in literature are typically distance based which measure the spatial distance between two feature vectors in high dimensional feature space. However, this type of similarity measures do not have any perceptual meaning and ignore the neighborhood influence in the similarity decision making process. In this paper, we propose a novel dissimilarity measure, which can measure both the distance and perceptual similarity of two image features in feature space. Results show the proposed similarity measure has a significant improvement over the traditional distance based similarity measure commonly used in literature.
Description:
International Conference Image and Vision Computing New Zealand
Description:
Previous empirical works have shown the effectiveness of differential prioritization in feature selection prior to molecular classification. We now propose to determine the theoretical basis for the concept of differential prioritization through mathematical analyses of the characteristics of predictor sets found using different values of the DDP (degree of differential prioritization) from realistic toy datasets. Mathematical analyses based on analytical measures such as distance between classes are implemented on these predictor sets. We demonstrate that the optimal value of the DDP is capable of forming a predictor set which consists of classes of features which are well separated and are highly correlated to the target classes – a characteristic of a truly optimal predictor set. From these analyses, the necessity of adjusting the DDP based on the dataset of interest is confirmed in a mathematical manner, indicating that the DDP-based feature selection technique is superior to both simplistic rank-based selection and state-of-the-art equal-priorities scoring methods. Applying similar analyses to real-life multiclass microarray datasets, we obtain further proof of the theoretical significance of the DDP for practical applications
Description:
The traditional way of using Hough Transform with SIFT is for the purpose of reliable object recognition. However, it cannot be effectively applied to image registration in the same way as the recall rate can be significantly lower. In this paper, we propose an alternative implementation of Hough Transform that can be used with Improved Symmetric-SIFT for multi-modal image registration. Our experimental results show that the proposed technique of applying Hough Transform can significantly improve the key-point matching as well as registration accuracy by utilizing aggregated information from key-points throughout the input images.
Description:
Corner detection is an important operation in many computer vision applications. Among the contour-based corner detectors in the literature, the Chord-to-Point Distance Accumulation (CPDA) detector is reported to have one of the best repeatability and lowest localization error. However, we found that CPDA detector often fails to accurately detect the true corners in some situations. Furthermore, CPDA detector is also computationally expensive. To overcome these weaknesses of CPDA detector, we propose an effective but yet efficient corner detector using a simple triangular theory. Our experimental results show that our proposed detector outperforms CPDA and six other existing detectors in terms of repeatability. Our proposed detector also has one of the lowest localization error. Finally it is computationally the most efficient.
Description:
To achieve scale-invariance, the approach used by many corner detection and description methods is to derive an appropriate scale as part of the process of detecting each corner and then use this scale for estimating region(s) around the corner to build the descriptor(s). However, this approach is not suitable for methods that do not derive such scale information in their corner detection process. This paper proposes a new method for selecting regions around a corner so that descriptors, which are invariant to scale changes and other image transformations, can be built to represent the corner. Our experimental results show that our proposed method achieves better precision-and-recall results than existing methods.
Description:
Symmetric-SIFT is a recently proposed local technique used for registering multimodal images. It is based on a well-known general image registration technique named Scale Invariant Feature Transform (SIFT). Symmetric SIFT makes use of the gradient magnitude information at the image's key regions to build the descriptors. In this paper, we highlight an issue with how the magnitude information is used in this process. This issue may result in similar descriptors being built to represent regions in images that are visually different. To address this issue, we have proposed two new strategies for weighting the descriptors. Our experimental results show that Symmetric-SIFT descriptors built using our proposed strategies can lead to better registration accuracy than descriptors built using the original Symmetric-SIFT technique. The issue highlighted and the two strategies proposed are also applicable to the general SIFT technique.