Curvature Scale Space Features Research Articles

Computer aided detection of spina bifida using nearest neighbor classification with curvature scale space features of fetal skulls extracted from ultrasound images

This paper addresses the problem of detecting the common neural tube defect of spina bifida by a computer aided detection (CAD) system. We propose a method which extracts the curvature scale space (CSS) features of fetal skull contours viewed in the ultrasound (US) modality and performs nearest neighbor (kNN) classification on those features having the desired properties of invariance with respect to translation, orientation and scale changes, thus improving robustness. The distance between two sets of CSS features, each set corresponding to the description of the contour of a particular skull, is measured as the cost of matching the two sets of CSS features. Such a CAD system may act as a second observer and help experts in prenatal diagnosis.Our data possess absolute and relative rarity. The experiments are performed with two different rare class handling methods and over a range of operating conditions. All experiments are based on a group of settings; associated with using either balanced or unbalanced datasets, employing different types of CSS features and how CSS matching costs are computed. Comparatively evaluating the classification performance of the settings is carried with the aid of the whole-curve metric of area under the receiver operating characteristics (ROC) curve (AUC). Optimal operating conditions for any setting can be identified and some settings reveal advantages over others. The observations indicate that using balanced datasets offers better performance and our proposed version of estimating CSS matching costs is generally superior to the classical method. Furthermore, using enhanced sets of CSS features improves classification accuracy. When classification is performed on balanced data using enhanced CSS features and the matching cost is computed with our proposed technique; one can observe an F-measure of 0.76 along with 70% TP rate (recall), 17% FP rate (false alarms) and 82% precision.

Shape feature encoding via Fisher Vector for efficient fall detection in depth-videos

Elderly people, who are living alone, are at great risk if a fall event occurred. Thus, automatic fall detection systems are in demand. Some of the early automatic fall detection systems such as wearable devices has a high cost and may cause inconvenience to the daily lives of the elderly people. In this paper, an improved depth-based fall detection system is presented. Our approach uses shape based fall characterization and a Support Vector Machines (SVM) classifier to classify falls from other daily actions. Shape based fall characterization is carried out with Curvature Scale Space (CSS) features and Fisher Vector (FV) encoding. FV encoding is used because it has several advantages against the Bag-of-Words (BoW) model. FV representation is robust and performs well even with simple linear classifiers. Extensive experiments on SDUFall dataset, which contains five daily activities and intentional falls from 20 subjects, show that encoding CSS features with FV encoding and a SVM classifier can achieve an up to 88.83% fall detection accuracy with a single depth camera. This classification rate is 2% more accurate than the compared approach. Moreover, an overall 64.67% accuracy is obtained for 6-class action recognition, which is about 10% more accurate than the compared approach.

A novel framework for Farsi and latin script identification and Farsi handwritten digit recognition

Optical character recognition is an important task for converting handwritten and printed documents to digital format. In multilingual systems, a necessary process before OCR algorithm is script identification. In this paper novel methods for the script language identification and the recognition of Farsi handwritten digits are proposed. Our method for script identification is based on curvature scale space features. The proposed features are rotation and scale invariant and can be used to identify scripts with different fonts. We assumed that the bilingual scripts may have Farsi and English words and characters together; therefore the algorithm is designed to be able to recognize scripts in the connected components level. The output of the recognition is then generalized to word, line and page levels. We used cluster based weighted support vector machine for the classification and recognition of Farsi handwritten digits that is reasonably robust against rotation and scaling. The algorithm extracts the required features using principle component analysis (PCA) and linear discrimination analysis (LDA) algorithms. The extracted features are then classified using a new classification algorithm called cluster based weighted SVM (CBWSVM). The experimental results showed the promise of the algorithms.

Feature alignment approach for hand posture recognition based on curvature scale space

One of the most important aspects of gesture recognition is recognizing hand postures. Much research has been devoted to extracting reliable features for hand posture recognition. In this paper, a new feature alignment approach for hand posture recognition based on curvature scale space (CSS) is presented. The basis point for alignment is based on the two-dimensional distribution of a coordinate-peak set of the CSS image instead of on the coordinate with the maximal peak. A convolution operation is performed with the sequence of a coordinate-peak set and a predefined function. The coordinate with the maximal convolution value is designated as a basis point for aligning the CSS features of the hand posture. Results show that the proposed approach performs well in recognizing hand postures. Furthermore, the proposed approach is more accurate than previous methods based on conventional features.

Feature-preserving algorithm for gesture recognition

This paper presents an approach for gesture recognition based on curvature scale space (CSS). First, for each gesture sequence, hands are segmented from gesture images into binary silhouette images, and then the binary hand contours are computed. Next, CSS features of hand contours and four-directional chain codes between two successive hand contours are extracted to characterize hand shapes and trajectories of gestures, respectively. In particular, a feature-preserving algorithm is proposed to allocate the extracted CSS features to a one-dimensional feature vector with a fixed size for the input of hidden Markov models (HMMs). Finally, the HMM is applied to determine hand-shape and trajectory transitions in different gestures for gesture identification. Results show that the proposed approach performs well in recognizing the gestures and is more accurate than the previous methods that were based on conventional features.

Adaptive multiple sets of CSS features for hand posture recognition

In this paper, an adaptive feature extraction approach based on curvature scale space (CSS) is presented for translation, scale, and rotation invariant recognition of hand postures. First, hands are segmented from hand posture images into binary silhouettes and then binary hand contours are computed. CSS images are then used to represent the contours of hand postures. In particular, adaptive multiple sets of CSS features are extracted to address the problem of deep concavities in the contours of hand postures. Finally, 1-nearest neighbor techniques are used to perform adaptive multiple sets of CSS feature matching for hand posture identification. Results indicate that the proposed approach performs well in the recognition of hand postures. And, the proposed approach is more accurate than previous methods which were based on conventional features. The proposed technique could be useful in improving the recognition of hand postures.