Search

A lot of useful data on forest condition can be gathered from the Forest Inven­tory (FI). Without the help of data analysis tools, human experts cannot ma­nually interpret information in such a large data set. Conventional multivariate statistical analyses provide results that are difficult to interpret and often do not represent the information in a satisfactory way. Our goal is to identify an alternative approach that will enable fast and efficient interpretation and analysis of the FI data. Such interpretation and analysis can be performed automatically with a clustering method, but all clustering methods have some shortcomings. Therefore, our aim was also to provide information in a form suitable for fast and intuitive visualization. Kohonen’s Self Organizing Map (SOM) is an alternative approach to data visualization and analysis of large multidimensional data sets. SOM provides different possibilities and our experiments are presented with component matrices of individual stand parameters and label matrices. In forming data clusters, we experimented with hierarchi­cal and non hierarchical clustering methods. Our experiments showed that SOM provides useful information in a form suitable for data clustering and data vi­sualization. This enables an efficient analysis of large FI data sets at different analysis scales. Clustering results obtained with SOM and two clustering algorithms are in accordance with ground truth. We have also considered the efficiency of SOM component matrices by visual comparison and correlation among structural parameters and by determining contributions of individual stand parameters to clustering input data. SOM application in visualization and analysis of stand structural parameters enables gathering quickly and efficiently holistic information on the current condition of forest stands and forest ecosystem development. Therefore we recommend the application of Kohonen’s SOM for visualization and analysis of FI data.

Nondestructive evaluation (NDE) is a set of techniques used for material inspection and defect detection without causing damage to the inspected component. One of the commonly used nondestructive techniques is called ultrasonic inspection. The acquisition of ultrasonic data was mostly automated in recent years, but the analysis of the collected data is still performed manually. This process is thus very expensive, inconsistent, and prone to human errors. An automated system would significantly increase the efficiency of analysis, but the methods presented so far fail to generalize well on new cases and are not used in real-life inspection. Many of the similar data analysis problems were recently tackled by deep learning methods. This approach outperforms classical methods but requires lots of training data, which is difficult to obtain in the NDE domain. In this work, we train a deep learning architecture EfficientDet to automatically detect defects from ultrasonic images. We showed how some of the hyperparameters can be tweaked in order to improve the detection of defects with extreme aspect ratios that are common in ultrasonic images. The proposed object detector was trained on the largest dataset of ultrasonic images that was so far seen in the literature. In order to collect the dataset, six steel blocks containing 68 defects were scanned with a phased-array probe. More than 4000 VC-B-scans were acquired and used for training and evaluation of EfficientDet. The proposed model achieved 89.6% of mean average precision (mAP) during fivefold cross validation, which is a significant improvement compared to some similar methods that were previously used for this task. A detailed performance overview for each of the folds revealed that EfficientDet-D0 successfully detects all of the defects present in the inspected material.

Two young female dogs were presented with a chronic history of persistent vulvar hemorrhage. Vaginoscopy was ultimately used to locate bleeding vaginal masses near the urethral papilla. In both cases, episiotomy was performed to resect the mass, and hemangioma was diagnosed histologically. These tumors caused persistent vaginal bleeding; they were difficult to diagnose without vaginoscopy and lavage; and surgical excision was curative in at least one case.

In this paper we propose a technique for 3-D segmentation of abdominal aortic aneurysm (AAA) from computed tomography (CT) angiography images. Output data form the proposed method can be used for measurement of aortic shape and dimensions. Knowledge of aortic shape and size is very important for selection of appropriate stent graft device for treatment of AAA. The technique is based on a 3-D deformable model and utilizes the level-set algorithm for implementation of the method. The method performs 3-D segmentation of CT images and extracts a 3-D model of aortic wall. Once the 3-D model of aortic wall is available it is easy to perform all required measurements for appropriate stent graft selection. The method proposed in this paper uses the level-set algorithm instead of the classical active contour algorithm developed by Kass et al. The main advantage of the level set algorithm is that it enables easy segmentation surpassing most of the drawbacks of the classical approach. In the level-set approach for shape modeling, a 3-D surface is represented by a real 3-D function or equivalent 4-D surface. The 4-D surface is then evolved through an iterative process of solving the differential equation of surface motion. Surface motion is defined by velocity at each point. The velocity is a sum of constant velocity and curvature-dependent velocity. The stopping criterion is calculated based on image gradient. The algorithm has been implemented in MATLAB and C languages. Experiments have been performed using real patient CT angiography images and have shown good results.