Spatial Point-Data Reduction Using Pulse Coupled Neural Network

2010 ◽  
Vol 32 (1) ◽  
pp. 11-29 ◽  
Author(s):  
Yongsheng Sang ◽  
Zhang Yi ◽  
Jiliu Zhou
Keyword(s):  
Neural Network ◽  
Data Reduction ◽  
Pulse Coupled Neural Network ◽  
Spatial Point ◽  
Point Data

scholarly journals Mapping of slopes for the operation of agricultural harvesters in Bandeirantes Municipality (PR)

2017 ◽  
Vol 38 (1) ◽  
pp. 97
Author(s):  
Gustavo Rodrigues Gimenes ◽  
Rone Batista Oliveira ◽  
Alessandra Fagioli da Silva ◽  
Luiz Carlos Reis ◽  
Teresinha Esteves da Silveira Reis
Keyword(s):  
Ground Surface ◽  
Digital Terrain Model ◽  
Large Area ◽  
Terrain Model ◽  
Spatial Point ◽  
Aster Gdem ◽  
Digital Terrain ◽  
Contour Lines ◽  
High Resolution Images ◽  
Point Data

The slope of terrain represents a risk factor for mechanized harvesting, leading to impediments or restrictions on agricultural operations, or even to machines toppling over in the field. Recently, the Digital Terrain Model (DTM) has become widely adopted as one of the most viable techniques for obtaining slope and elevation. Therefore, this study aims to assess methods of acquiring DTMs to calculate the slope, and to determine the areas that are suitable and unsuitable for the operation of harvesters in the municipality of Bandeirantes (PR). Four methods were selected to produce DTMs for the construction of slope zoning maps applicable for harvester operations. The image sources included SRTM, ASTER GDEM, digitizing contour lines and kriging of spatial point data. After generating DTMs by the four different methods, the area suitable for the operation of harvesters was obtained based on the limits of operational slopes for harvesters in the literature. The high-resolution images, such as those obtained by scanning the contour lines and ASTER GDEM gave the best representation of the ground surface. Regardless of the method used to obtain the operational slopes, the municipality has a large area that is suitable for mechanized harvesting.

scholarly journals A New Retrieval Algorithm Based on Pulse Coupled Neural Network for Biomedical Images

2019 ◽  
Vol 9 (1S) ◽  
pp. 9-13
Keyword(s):  
Neural Network ◽  
Magnetic Resonance Imaging ◽  
Magnetic Resonance ◽  
Rapid Expansion ◽  
Image Feature ◽  
Experimental Result ◽  
Feature Descriptor ◽  
Resonance Imaging ◽  
Biomedical Images ◽  
Pulse Coupled Neural Network

The rapid expansion and improvement in medical science and technology lead to the generation of more image data in its regular activity such as computed tomography (CT), X-ray, magnetic resonance imaging (MRI) etc. To manage the medical images properly for clinical decision making, content-based medical image retrieval (CBMIR) system emerged. In this paper, Pulse Coupled Neural Network (PCNN) based feature descriptor is proposed for retrieval of biomedical images. Time series is used as an image feature which contains the entire information of the feature, based on which the similar biomedical images are retrieved in our work. Here, the physician can point out the disorder present in the patient report by retrieving the most similar report from related reference reports. Open Access Series of Imaging Studies (OASIS) magnetic resonance imaging dataset is used for the evaluation of the proposed approach. The experimental result of the proposed system shows that the retrieval efficiency is better than the other existing systems.

Sign in / Sign up

Export Citation Format

Share Document