Pattern Recognition and Image Analysis

This paper reports a novel region-based shape descriptor based on orthogonal Legendre moments. The preprocessing steps for invariance improvement of the proposed Improved Legendre Moment Descriptor (ILMD) are discussed. The performance of the ILMD is compared to the MPEG-7 approved region shape descriptor, angular radial transformation descriptor (ARTD), and the widely used Zernike moment descriptor (ZMD). Set B of the MPEG-7 CE-1 contour database and all the datasets of the MPEG-7 CE-2 region database were used for experimental validation. The average normalized modified retrieval rate (ANMRR) and precision-recall pair were employed for benchmarking the performance of the candidate descriptors. The ILMD has lower ANMRR values than ARTD for most of the datasets, and ARTD has a lower value compared to ZMD. This indicates that overall performance of the ILMD is better than that of ARTD and ZMD. This result is confirmed by the precision-recall test where ILMD was found to have better precision rates for most of the datasets tested. Besides retrieval accuracy, ILMD is more compact than ARTD and ZMD. The descriptor proposed is useful as a generic shape descriptor for content-based image retrieval (CBIR) applications.

The ease in digital imaging has led to a decrease in image fidelity where illegal reproduction of multimedia information has become difficult to detect. The most challenging problem is to protect image copyright against illegal copies. Therefore a watermark detection process is required to verify the owner of the image. This paper proposes a blind algorithm to extract the watermark. This algorithm is robust against noise and geometric attacks for grayscale images. Robust feature points are detected using the Harris Detector. In the embedding stage, sequences are placed into regions located around feature points. The sequences used are PN, Gold and decimal. In the extraction process, image features are re-allocated using the same detector. Each feature point is used as a center of an N × N region. This region is moved horizontally and vertically within its neighboring pixels. Each move should be registered in a matrix as a correlation value of this region with the initial sequence. This procedure is repeated for all feature points till we find all the watermarked regions. The maximum correlation obtained will determine the center of the watermarked region. The proposed algorithm is robust against a wide variety of tests and is compared to other schemes.

We present a combinatorial deep convolutional neural network architecture, termed strong convolution neural network (SSN), for semantic segmentation task. The structure of SSN consists of two components: Increment feature convolution neural network and post-process Conditional Random Fields unit (CRFs). The increment feature CNN unit has three parts: I-Block, Deconvolution layer and Transition Block. I-Block employs increment convolution to efficiently maintain feature information. Before passing through pooling layer, we put the feature map into activate layer ReLU, and batch normalization layer. In Decoding stage, we use skip-connects to keep the pooling index information. To enforce the correlation of same semantic labels, we define the strong semantic label (SSL) stage to intensify the pairwise potential energy. To achieve high computation performance, we make further improvement on SSL by employing the adaptive soft semantic sections label method. We proposed the adaptive strong semantic label selection algorithm to generate the SSL. Through the CRFs unit, with unitary energy and pairwise edge energy, the semantic segmentation initial labels transform semantic segmentation labels. Experimental evaluation reveals the training time versus accuracy trade-off involved in achieving good segmentation performance.

In ophthalmology, blood vessel segmentation from fundus images plays a significant role in automated retinal disease screening systems. Several research papers on blood vessel segmentation suggest enhancing fundus images before segmentation significantly to improve performance. The brightness of the optic disc region in a fundus image negatively influences the enhancement of relatively darker vessel pixels. Segregation of brighter optic disc from fundus images before its enhancement is the fundamental idea behind developing the proposed framework. Initially, the optic disc is extracted from the input fundus image to form two images, one containing optical disc and the other, fundus image without optical disk. In the second stage, both the images are enhanced independently, followed by blood vessel segmentation. Finally, the segmented blood vessels from the images are fused to obtain a single image. Experiments conducted with fundus images from DRIVE, STARE, and CHASE_DB1 databases show improvement in the identification of blood vessel pixels.

The paper summarizes the joint work of specialists in the fields of image analysis and ophthalmology over the last few years. As a result of this work, new mathematical methods for automating image analysis that have important diagnostic value for ophthalmology have been developed: (1) identification of the lipid layer state in the intermarginal space of human eyelids; (2) analysis of the degree of cellular structure density (cellularity) in the corneal tissue of human eyes; (3) identification of the state of the retinal blood flow when analyzing fluorescent angiograms of the human fundus; and (4) morphometric analysis of the state of the epithelium posterius (endothelium) in the human eye cornea. As initial data, we used (respectively) (1) images of imprints of the eyelid intermarginal space on a millipore filter upon their osmium vapor staining; (2) micrographs of corneal tissue specimens obtained using a light microscope; (3) fluorescent angiograms of the human fundus; and (4) images of endothelial cells obtained noninvasively using a confocal microscope. The developed methods are designed to extract morphometric data from these images. For each problem, the following results were obtained: (1) expectations and variances of pixel intensities on the imprint along a drawn line and over a selected region, as well as plots that characterize pixel intensity and change in the thickness of the imprint along a drawn line; (2) expectations and variances for the intensities of the selected regions and intensity histograms; (3) extracted vessels and ischemia zones with their statistical descriptions; and (4) detected cells of hexagonal, pentagonal, and other shapes, as well as a set of characteristics associated with the size of the cells detected. The developed methods are based on the fundamental results of the mathematical theory of image analysis and on the joint use of image processing, mathematical morphology, and mathematical statistics techniques. The paper also describes software implementations of the developed methods, including an automated research workstation for ophthalmologists, and presents the results of their experimental testing.

The problem of extracting statistical data from a table of a basic structure is considered. The problem of extracting statistical indicators is formally stated. The method for extracting statistical data that analyzes the content of table cells is described. Algorithms for cell type recognition, table structure restoration, and a fuzzy search of the content of text cells are proposed.

We consider the problem of increasing the productivity of three-dimensional model restoration methods. Issues relating to the use of multicore and multiprocessor systems are discussed. The architecture of the latest computer systems that use GPU as multicore high-speed calculators is also considered, and aspects of their use to increase the performance of stereo processing methods are also considered.

A no-reference method to detect and suppress ringing effect in MRI images is suggested. The ringing detection method is based on finding the area where ringing effect is likely to appear and calculating the ratio of average edge-normal and edge-tangential derivatives moduli in this area. The area consists of pixels with the certain distance to basic edges—sharp edges that are distant from other edges. The proposed ringing suppression method is based on the projection onto the set of images with bounded total variation with ringing level control.

Ridges (resp., valley) are the useful geometric features due to their wide varieties of applications, mainly in image analysis problems such as object detection, image segmentation, scene understanding, etc. Many characterizations have contributed to formalize the ridge notion. The signification of each characterization rely however on its actual application. The objective analysis of ridge characterized as thin and complex image structure is thus essentially important, for choosing which parameter’s values correspond to the suitable configuration to obtain accurate results and optimal performance. In this article an extensive analysis followed by a supervised and objective comparison of different filtering-based ridge detection techniques is led. Furthermore, the optimal parameter configuration of each filtering techniques aimed for image salient feature analysis tool have been objectively investigated, where each chosen filter’s parameters corresponds to the width of the desired ridge or valley. At last, the comparative evaluations and analysis results are reported on both synthetic images, distorted with various types of noises and real images.

A framework is proposed for fast calculation of linear scalings posed on structured data. Several widely used types of data representation by clusters with intrinsic features of local simmetry are taken into account. Paper presents some Image Mining technologies that are used for improvement of abstract data multiview evaluation procedures.