Search

Inversion of electromagnetic induction data is often realized using smoothing operators for regularizing the process. These operators are applied using derivative matrices, penalizing differences between neighboring mesh cells. The resulting smoothing effect is similar to that of low-pass Gaussian filters and may cause difficulties due to the smoothed boundaries between structures. To prevent such ambiguities, non-linear rank filters are applied widely in image processing applications to filter low frequency variations while keeping boundaries. Correspondingly, we have defined a non-linear rank order smoothing operator for regularization. The effects of the defined parameters are demonstrated on synthetic models and the operator is observed to be able to provide better defined boundaries. The implementation of the operator on field data is realized using single-frequency VLF-R data collected on profiles traversing the surface rupture of the North Anatolian Fault in Kocaeli, Turkey. In the study area, ERT data are also collected on a single profile, and its smooth inversions are realized individually and jointly with VLF-R data for comparisons. Inversions show that the recovered subsurface structure using the rank order smoothing is closer to that indicated by the ERT data and previous studies. According to the recovered models, the clayey fillings in the fault rupture are acting as a barrier affecting the horizontal fluid flow in the near-surface and the resulting water accumulations around the fault rupture may cause susceptibility for earthquake induced damages. • VLF-R data is inverted using a smoothing operator based on rank order filters. • The operator is observed to be able to provide models with better defined boundaries. • VLF-R data collected on the rupture of the North Anatolian Fault is inverted. • Inversions suggest, the clayey fills in the rupture causes water accumulations around the fault. [ABSTRACT FROM AUTHOR]

The need for timely, accurate, and interoperable geospatial information is steadily increasing. In this context, process-based image processing systems will be the initial segment for future’s automatic systems. A process-based system is believed to be a good approach for agricultural purpose because agricultural activities are carried out according to a periodic (annual) cycle. Therefore, a process-based image analysis procedure was designed for routine crop classification for an agricultural region in Kırklareli, Turkey. The process tree developed uses a multi-temporal image data set as an input and gives the final crop classification as an output by using an incremental rule set. The test data set was composed of five images of Satellite Pour l’Observation de la Terre 4 (SPOT 4) data acquired in 2007. Basically, image objects were first extracted and then classified. A rule set was structured depending on class definitions. A decision-based process was executed and formed a multilevel image classification system. The final classification was obtained by merging classes from the appropriate levels where they were extracted. To evaluate the success of the application the accuracy of the classification was assessed. The overall accuracy and kappa index of agreement was found to be 80% and 0.78, respectively. At the end of the study, problems of segmentation and classification operations were discussed and solution approaches were outlined. To assess the process in terms of its scope for automation, the efficiency and success of the rule set were also discussed. [ABSTRACT FROM AUTHOR]

In this study, supplementary information on the morphology of the siphonostomatoid copepod Caligus quadratus Shiino, 1954 (Copepoda: Caligidae) is given based on the new material collected from the gills of the Atlantic bluefin tuna, Thunnus thynnus (Linnaeus, 1758) caught in the Gulf of Antalya, Turkey. The morphology of C. quadratus is re-examined for the first time by adopting a recently developed visualisation technique by Kamanli et al. [1]. Appendages of Congo red stained specimens of C. quadratus were dissected and scanned using confocal laser scanning microscope (CLSM) and the CLSM images were processed using Dristhi software programme. 3D reconstructions of some confusing appendages were visualised using Drishti. Line drawing was used to depict the habitus of both female and male C. quadratus. Key diagnostic characters of C. quadratus are presented together with the newly observed additional taxonomic characters. In addition, previously misinterpreted and simply overlooked features in the previous descriptions of C. quadratus are also re-described. This is the first report of C. quadratus from the Mediterranean Sea and the Atlantic bluefin tuna constitutes a new host record for this caligid copepod. • Atlantic bluefin tuna constitutes a new host record for Caligus quadratus. • It is the first record of Caligus quadratus in the Mediterranean Sea. • Supplementary information on the morphology of Caligus quadratus were presented. • A new technique, CLSM and Drishti 3D software, was used to describe C. quadratus. [ABSTRACT FROM AUTHOR]