Direction Of Intensity Gradient Research Articles

Hough Transform Application to Digitize Rectangular Spatial Objects on Aerospace Imagery

The paper describes the method of development for the remote sensing data processing to speed up the digitizing workflow. The method is designed to digitize rectangular objects using their approximate spatial positions and provides an automatic estimation of the orientation and aspect ratio. 
 The paper contains a formal statement of the problem of digitizing an object with the desired geometric shape using it’s apriori known spatial position on a source image. The method creates polygonal representations of rectangular spatial objects from one or a few reference points set by an operator. It is based on source image’s pixels clustering using spectral bands as a feature space. The following Hough transform incorporates local direction of intensity gradient to estimate object’s orientation and reduce computational complexity together with low-pass filtering within an accumulation process to improve robustness. It is shown that the developed method can be modified to digitize objects of any analytically described shape. 
 The method is designed to allow easy user interaction without any significant delays and to provide transparent and predictable control of an output object’s polygon size.
 To investigate the developed method a test dataset with more than 700 rectangular objects was used. The root-mean-square error of object’s points positioning, mean rotation error in polar coordinates and a Jaccard index were used to measure a precision of the digitized objects. The experiment results demonstrate that digitizing workflow is accelerated by 25–40% using the software implementing the developed method without a significant precision loss.

Detection of Magnetic Field Intensity Gradient by Homing Pigeons (Columba livia) in a Novel “Virtual Magnetic Map” Conditioning Paradigm

It has long been thought that birds may use the Earth's magnetic field not only as a compass for direction finding, but that it could also provide spatial information for position determination analogous to a map during navigation. Since magnetic field intensity varies systematically with latitude and theoretically could also provide longitudinal information during position determination, birds using a magnetic map should be able to discriminate magnetic field intensity cues in the laboratory. Here we demonstrate a novel behavioural paradigm requiring homing pigeons to identify the direction of a magnetic field intensity gradient in a “virtual magnetic map” during a spatial conditioning task. Not only were the pigeons able to detect the direction of the intensity gradient, but they were even able to discriminate upward versus downward movement on the gradient by differentiating between increasing and decreasing intensity values. Furthermore, the pigeons typically spent more than half of the 15 second sampling period in front of the feeder associated with the rewarded gradient direction indicating that they required only several seconds to make the correct choice. Our results therefore demonstrate for the first time that pigeons not only can detect the presence and absence of magnetic anomalies, as previous studies had shown, but are even able to detect and respond to changes in magnetic field intensity alone, including the directionality of such changes, in the context of spatial orientation within an experimental arena. This opens up the possibility for systematic and detailed studies of how pigeons could use magnetic intensity cues during position determination as well as how intensity is perceived and where it is processed in the brain.

Ensemble Systems for Automatic Fracture Detection

Fracture detection based on image classification is an area of research which has proved to be challenging for the past several decades. This field has gained more attention due to the new challenges posed by voluminous image databases. In this research work, fusion-based classifiers are constructed, which extracts features from the images, use these features to train and test the classifiers for the purpose of detecting fractures in X-Ray images. The various features extracted are Contrast, Homogeneity, Energy, Entropy, Mean, Variance, Standard Deviation, Correlation, Gabor orientation (GO), Markov Random Field (MRF), and intensity gradient direction (IGD). Three classifiers, BPNN, SVM and NB classifiers are used. Using these features and classifiers, three single classifiers and four multiple classifiers were developed. All the classifiers were tested vigorously with the test dataset for evaluating the winner combination of classifiers and features that correctly identifies fractures in a bone image. The performance metrics used are sensitivity, specificity, positive predictive value, negative predictive value, accuracy and execution time. The experimental results showed that usage of fusion classifiers enhances the detection capacity and the combination SVM and BPNN produces the best result.

Multiple feature sets based categorization of laryngeal images

This paper is concerned with an automated analysis of laryngeal images aiming to categorize the images into three decision classes, namely healthy, nodular, and diffuse. The problem is treated as an image analysis and classification task. Aiming to obtain a comprehensive description of laryngeal images, multiple feature sets exploiting information on image colour, texture, geometry, image intensity gradient direction, and frequency content are extracted. A separate support vector machine (SVM) is used to categorize features of each type into the decision classes. The final image categorization is then obtained based on the decisions provided by a committee of support vector machines. Bearing in mind a high similarity of the decision classes, the correct classification rate of over 94% obtained when testing the system on 785 laryngeal images recorded at the Department of Otolaryngology, Kaunas University of Medicine is rather promising.

Motion estimation based on the direction of intensity gradient

The optical flow constraint (OFC) equation has been extensively studied in computer vision to estimate motion from image sequences. Traditionally the OFC relies on spatio-temporal brightness variations caused by motion. However, many other features are in principle appropriate, including gradient directions of the image intensity. Given gradient directions are more tolerant to changes in lighting, they seem to be an adequate choice for computing optical flows. Their applicability in the OFC is, however, not straightforward as the gradient direction is independent of the gradient magnitude, and is not defined in homogeneous areas. To palliate to these difficulties, a form of OFC equation based on probability distributions of gradient directions is proposed. The performance of the approach is assessed by experiments realized on synthetic and real world images.

Automatic analysis of electron backscatter diffraction patterns

The ability to measure lattice orientation in individual crystallites enables a more complete characterization of microstructure by combining lattice orientation with morphological features. Lattice orientation can be obtained by analyzing electron backscatter diffraction patterns (EBSPs). However, current computer-aided EBSP analysis techniques make it impractical to obtain the number of measurements needed for statistically reliable characterizations of microstructure. An effective fully automated technique for determining crystallographic orientation from EBSPs is described. Bands are identified by linear regions of correlation in the image intensity gradient direction. The most probable orientation is then found using the angles between the detected bands. The reliability of the technique was tested using a set of 1000 patterns obtained from annealed oxygen-free electrical grade (OFEC) copper. The orientation of each test pattern found using automatic indexing was checked against the corresponding orientation as determined by manual indexing. Ninety-nine percent of auto-indexed orientations were found to lie within 5 deg of the misorientation angle of the manual-indexed orientations. By simulating noise in the test patterns, it was found that image quality has a strong effect on the reliability of the technique. An image quality parameter is described which allows the reliability of the technique to be predicted for a pattern of given quality.