Rituparna Sarkar

Dictionary Learning Level Set

We propose a novel region based segmentation technique using dictionary learning. In a previous work we have developed a method which uses a set of pre-specified Legendre basis functions to perform region based segmentation of an object in presence of heterogeneous illumination. We hypothesize that in problems where a set of training images for the object is available for analysis (such as depth image sequence of blood vessels via ultrasound imaging), segmentation accuracy can be significantly improved by learning the basis functions instead of specifying them implicitly. The salient idea of this letter is to compute the optimal set of functions to model the region intensities. Our solution to this problem involves the integration of a level set segmentation methodology with the dictionary learning framework. This provides an elegant solution to deal with intensity inhomogeneities prevalent in many imaging applications such as ultrasound and fluorescence microscopy. The proposed algorithm, Dictionary Learning Level Set (DL2S) is used to segment ultrasound images of blood vessels captured using low cost, portable ultrasound devices employed in a phlebotomy application. Qualitative and quantitative results obtained from this dataset suggest efficacy of D2LS with an associated improvement in the average Dice index of 12% over the relevant competitors.

Image classification by multi-kernel dictionary learning

Recent studies have indicated the efficacy of selecting and combining the salient features from a pool of feature types in image retrieval and classification applications. In contrast to previous work, in this paper, we approach this problem as a selection and combination of the salient feature type(s) from a pool of feature types rather than selecting an individual feature. Our approach utilizes multiple kernels within the dictionary-learning framework where a combination of dictionary atoms represents individual categories. The category specific feature combination parameters or weights for kernel combination are determined by the mutual information techniques. The method is compared to a meta-algorithm for feature nomination. The multi-kernel dictionary learning method yields, on average, a 10% increase in classification accuracy with respect to the meta-algorithm in our preliminary experiments.

Slide: Saliency guided image dictionary and image similarity evaluation

In this paper we present a novel idea of evaluating similarity between two images aided by a salient object detection framework. For computing similarity between images consisting of multiple objects and varying background, extracting features relevant to the object of interest is of cruicial importance. To accomplish this task, we employ a saliency guided dictionary learning framework for image similarity evaluation (SLIDE). The saliency detection framework emphasizes the image regions that attract human attention and is exploited to build a dictionary for generating sparse representation of the images. The compressibility of the sparse codes is exploited in computing the similarity measure. The SLIDE framework is used in the application of image retrieval for three military image datasets and shows in average an improvement of 14.8% in retrieval accuracy compared to two state-of-the-art similarity measures - information distance and sparse representation based compression distance.

Feature extraction and image retrieval on an automata structure

An automata processor can execute pattern matching in parallel which brings potential to accelerate image recognition. In this paper, we present a novel process of implementing image retrieval using multinary representation for use on an automata framework. Images are encoded into discriminative and unique regular expression descriptors in such a way that can be used for classification purposes. The regular expression descriptors are streamed through sets of non-deterministic finite automata (NFA). Results show that image retrieval can be implemented on automata structures that can achieve dramatic improvement over general purpose processors or graphics processors in efficiency.

A meta-algorithm for classification by feature nomination

With increasing complexity of the dataset it becomes impractical to use a single feature to characterize all constituent images. In this paper we describe a method that will automatically select the appropriate image features that are relevant and efficacious for classification, without requiring modifications to the feature extracting methods or the classification algorithm. We first describe a method for designing class distinctive dictionaries using a dictionary learning technique, which yields class specific sparse codes and a linear classifier parameter. Then, we apply information theoretic measures to obtain the more informative feature relevant to a test image and use only that feature to obtain final classification results. With at least one of the features classifying the query accurately, our algorithm chooses the correct feature in 88.9% of the trials.

Shape descriptors based on compressed sensing with application to neuron matching

In this paper we propose a novel compressed sensing based Fourier shape descriptor method to compute the shape feature vector of an arbitrary object. First, the object contour obtained via segmentation is represented as a complex-valued signal. We then formulate an optimization problem that exploits the sparsity of the shape feature of the contour. This results in a reduced size feature vector, which can efficiently represent the shape of an object as illustrated by the reconstruction results. Appropriate for general shape retrieval problems, we demonstrate the efficacy of our algorithm by retrieving structurally similar neurons from a database. Currently, the representation and matching of neurons, given the heterogeneous nature of the neuronal morphology and the characteristically complex branching patterns, is an open problem. Retrieval of structurally similar neurons will potentially enable classification of neurons imaged. The retrieval results obtained using our method provide evidence of efficacy with a 27% improvement over Sholl analysis, which is a standard shape descriptor used in neuroscience.

SSPARED: Saliency and sparse code analysis for rare event detection in video

The problem of detecting rare and unusual events in video is critical to the analysis of large video datasets. Such events are identified as those occurrences within a sequence that cause a significant change in the scene. We propose to determine the significance of a frame, while preserving its compact representation, by introducing a saliency-driven dictionary learning technique. The derived sparse codes are then leveraged, together with the Kullback-Leibler divergence, in the design of a histogram-based metric that we use to evaluate the scene changes between consecutive frames. Our method, SSPARED, is compared with two state of the art methods for anomaly detection and shows significant improvement in detecting abnormal incidents and reduced false alarm generation.

Tracking sunflower circumnutation using affine parametric active contours

Study of the sunflower movement may reveal clues regarding unknown mechanisms that regulate periodicity and spatial complexity in plant growth and development. In this paper, we introduce an automated process to track circumnutation of sunflower seedlings. The objective is to track the leaves of the sunflower plant in a video captured by an overhead camera. The tracking method presented predicts the translated and rotated boundary in the subsequent frames by active contour models. A salient feature of our solution is a constraint on affine transformation between updates. The constrained affine active contours used in this paper exhibit improvement over other traditional active contour approaches, with the new method yielding error less than one percent in the tracked sunflower centroid position.

SDL: Saliency-Based Dictionary Learning Framework for Image Similarity

In image classification, obtaining adequate data to learn a robust classifier has often proven to be difficult in several scenarios. Classification of histological tissue images for health care analysis is a notable application in this context due to the necessity of surgery, biopsy or autopsy. To adequately exploit limited training data in classification, we propose a saliency guided dictionary learning method and subsequently an image similarity technique for histo-pathological image classification. Salient object detection from images aids in the identification of discriminative image features. We leverage the saliency values for the local image regions to learn a dictionary and respective sparse codes for an image, such that the more salient features are reconstructed with smaller error. The dictionary learned from an image gives a compact representation of the image itself and is capable of representing images with similar content, with comparable sparse codes. We employ this idea to design a similarity measure between a pair of images, where local image features of one image, are encoded with the dictionary learned from the other and vice versa. To effectively utilize the learned dictionary, we take into account the contribution of each dictionary atom in the sparse codes to generate a global image representation for image comparison. The efficacy of the proposed method was evaluated using three tissue data sets that consist of mammalian kidney, lung and spleen tissue, breast cancer, and colon cancer tissue images. From the experiments, we observe that our methods outperform the state of the art with an increase of 14.2% in the average classification accuracy over all data sets.