SAR (synthetic aperture radar) and PolSAR (polarimetric synthetic aperture radar) images fulfill a fundamental role in Earth observation, due to their advantages over optical images. However, the presence of speckle noise hinders their automatic interpretation and unsupervised use, rendering traditional segmentation tools ineffective. For this reason, advanced image segmentation models are sought to overcome the limitations that make an adequate treatment of speckled images difficult. We propose a procedure for SAR and PolSAR image classification, based on texture descriptors, that combines fractal dimension, a specific probability distribution function, Tsallis entropy, and the entropic index. A vector of local texture features is built using a set of reference regions, then a support vector machine classifier is applied. The proposed algorithm is tested with synthetic and actual monopolarimetric and polarimetric SAR imagery, exhibiting visually remarkable and robust results in coincidence with quantitative quality metrics as accuracy and F1-score.
Image-based diagnosis is able to spot several diseases and clinical conditions faster and more accurately than traditional manual ones, becoming also an alternative in monitoring and predicting patients responses to specific health treatments. In this work, we present a supervised learning approach to segment pixel-wise parts of spermatozoa using a random forest (RF) classifier. The framework created a multi-channel image combining intensity RGB bands with three neighborhood based bands. The last neighborhood based bands were Sobel’s magnitude and orientation and Shannon’s entropy. A RF was trained using labeled pixels provided by expert andrologists, biochemists and specialists in reproductive health. We compared results with a simple model on the RGB only. The whole automatic process (segmentation and classification) achieved an average precision of 98%, recall of 98% and F-Score of 98%. Highest improvement in comparison to the RGB model was shown on the segmentation of the tail. We provided a fully automatic spermatozoa semantic segmentation based on local and non-local information. The results are aimed to develop a CASA (Computer Assisted Sperm Analysis) system that can provide results over the Internet. The experiment was conducted on normalized images of a specific microscope. We are planning to extend the experiment in future work to more realistic conditions including different stainings, microscopes and resolutions.
Image-based diagnosis becoming one of the most important areas in medicine, as the diversity and sophistication of imaging techniques are being increasingly used in hospitals and medical centers. This, however, raises the issue of having image analysis capabilities that go with this trend, to be able to use medical imagery to provide fast and accurate diagnosis. In andrology in particular, the spermiogram analysis is considered the most significant study to evaluate the male reproductive capacity. Spermiograms can be produced with relatively little effort and cost, since they require only standard procedures for sample treatment. However, an adequate assessment of sperm quality requires the careful inspection by higly trained specialists, requiring time, and being prone to high inter- and intra-specialist variances. In this paper we present a system for automatic spermiogram analysis using image processing and machine learning techniques. The system was trained using a repository of spermiograms and the opinion of several experts in andrology and in human reproduction, using different information sources and classification criteria. The results are aimed to develop a SaaS CASA (Computer Assisted Sperm Analysis) system that can provide results over the Internet.
Oil spillage is one of the most common sources of environmental damage in places where coastal wild life is found in natural reservoirs. This is especially the case in the Patagonian coast, with a littoral more than 5000 km long and a surface above a million and half square km. In addition, furtive fishery activities in Argentine waters are depleting the food supplies of several species, altering the ecological equilibrium. For this reason, early oil spills and vessel detection is an imperative surveillance task for environmental and governmental authorities. However, given the huge geographical extension, human assisted monitoring is unfeasible, and therefore real time remote sensing technologies are the only operative and economically feasible solution. In this work we describe the theoretical foundations and implementation details of a system specifically designed to take advantage of the SAR imagery delivered by two satellite constellations (the SAOCOM mission, developed by the Argentine Space Agency, and the COSMO mission, developed by the Italian Space Agency), to provide real-time detection of vessels and oil spills. The core of the system is based on pattern recognition over a statistical characterization of the texture patterns arising in the positive and negative conditions (i.e., vessel, oil, or plain sea surfaces). Training patterns were collected from a large number of previously reported contacts tagged by experts in the National Commission on Space Activities (CONAE). The resulting system performs well above the sensitivity and specificity of other avalilable systems.
This work is concerned with buried landmines detection by long wave infrared images obtained during the
heating or cooling of the soil and a segmentation process of the images. The segmentation process is performed by
means of a local fractal dimension analysis (LFD) as a feature descriptor. We use two different LFD estimators,
box-counting dimension (BC), and differential box counting dimension (DBC). These features are computed in
a per pixel basis, and the set of features is clusterized by means of the K-means method. This segmentation
technique produces outstanding results, with low computational cost.
The purpose of Scientific Visualization is to provide qualitative information by means of graphic methods, bringing quick and intuitive data interpretation. In stationary vector flows and non-linear and chaotic dynamic systems, visualization is specially useful due to the impossibility to find closed analytic solutions to the system's behavior. Numerous methods have been proposed for representing vector fields, of which streamlines an LIC are the most important. In streamlines, we perform a numerical evaluation of the flow by means of trajectories that are tangent to the vector field. This technique is adequate for real-time visualization, but has several drawbacks in accuracy. On the other hand, the technique known as LIC (line integral convolution) is a texture-based method, in which the trajectories originated at every pixel in the phase portrait, are advected along an input texture to find the final color of that pixel. The LIC overcomes most of the drawbacks of the streamlines, but is too slow to be useful as a real-time visualization tool. In this work we will show a new visualization method, called CLIC (cumulative line integral convolution), which effectively combines the advantages of streamlines and LIC. We further discuss the implementation of a real-time visualization tool that is adequate for bringing an Internet based visualization service.
Consider an hypothetical image processing system, where a given target is to be identified. The usual sequence of steps consists on an image equalization to adapt to the illumination situation. Then the image is binarized, allowing a morphological filter to correct the noisy edges and shapes by means of an indeterminate sequence of openings or closings. The resulting image can then be segmented and recognized. If the results are unsatisfactory, then the processing parameters in any of the previous steps must be changed, perhaps by trial and error. For instance, the binarization threshold can be raised or lowered, and the following steps must be performed again to see the results. This is obviously cumbersome, tedious and error prone. The Image Processing Spreadsheet PDICalc is a simple but powerful combination of two different and widespread software technologies. It's benefit comes from enabling users to build an image processing pipeline, considering each step separately, and visualizing the results of modifying the parameters of each step in the final image. A spreadsheet based user interface eliminates the tedious and repetitive interaction that characterizes current image processing software. Users can build a processing template and reliably repeat often needed processing without the effort of redevelopment or recoding. In the cited example the user simply creates the processing template, defining each cell of the spreadsheet as the result of applying a given processing step on another cell. This template can be then reused with any input image, can be stored for future processing sessions, and every step can be trimmed precisely to achieve the desired results. Our implementation considers most of the image processing techniques as its building blocks. Arithmetic operators are overloaded to represent per pixel operations. We included also equalization and histogram correction, arbitrary convolution filtering, arbitrary morphological filtering (with programmed repetition), Fourier operations, and several segmentation techniques.
Current research on artificial vision and pattern recognition tends to concentrate either on numerical processing (filtering, morphological, spectral) or in symbolic or subsymbolic processing (neural networks, fuzzy logic, knowledge-based systems). In this work we combine both kinds of processing in a hybrid image processing architecture. The numerical processing part implements the most usual facilities (equalization, convolution filters, morphological filters, segmentation and description) in a way adequate to transform the input image into a polygonal outline. Then recognition is performed with a rule-based system implemented in Prolog. This allows a neat high-level representation of the patterns to recognize as a set of logical relations (predicates), and also the recognition procedure is represented as a set of logical rules. To integrate the numerical and logical components of our system, we embedded a Prolog interpreter as a software component within a visual programming language. Thus, our architecture features both the speed and versatility of a visual language application, and the abstraction level and modularity of a logical description.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.