scholarly journals EL: Local Image Descriptor Based on Extreme Responses to Partial Derivatives of 2D Gaussian Function

2019 ◽  
Vol 2019 ◽  
pp. 1-10
Author(s):  
Jasna Maver ◽  
Danijel Skočaj
Keyword(s):  
Image Matching ◽  
Gaussian Function ◽  
Two Dimensional ◽  
Image Descriptor ◽  
Partial Derivatives ◽  
Illumination Changes ◽  
Local Image Descriptor ◽  
Extreme Responses ◽  
Derivatives Of ◽  
Local Image

We propose a two-part local image descriptor EL (Edges and Lines), based on the strongest image responses to the first- and second-order partial derivatives of the two-dimensional Gaussian function. Using the steering theorems, the proposed method finds the filter orientations giving the strongest image responses. The orientations are quantized, and the magnitudes of the image responses are histogrammed. Iterative adaptive thresholding of histogram values is then applied to normalize the histogram, thereby making the descriptor robust to nonlinear illumination changes. The two-part descriptor is empirically evaluated on the HPatches benchmark for three different tasks, namely, patch verification, image matching, and patch retrieval. The proposed EL descriptor outperforms the traditional descriptors such as SIFT and RootSIFT on all three evaluation tasks and the deep-learning-based descriptors DeepCompare, DeepDesc, and TFeat on the tasks of image matching and patch retrieval.

A local image descriptor based on radial and angular gradient intensity histogram for blurred image matching

2018 ◽  
Vol 35 (10) ◽  
pp. 1373-1391 ◽  
Author(s):  
Bahman Sadeghi ◽  
Kamal Jamshidi ◽  
Abbas Vafaei ◽  
S. Amirhassan Monadjemi
Keyword(s):  
Image Matching ◽  
Image Descriptor ◽  
Intensity Histogram ◽  
Blurred Image ◽  
Local Image Descriptor ◽  
Local Image

A new Rényi entropy-based local image descriptor for object recognition

2010 ◽  
Author(s):  
S. Gabarda ◽  
G. Cristóbal ◽  
P. Rodríguez ◽  
C. Miravet ◽  
J. M. del Cura
Keyword(s):  
Object Recognition ◽  
Renyi Entropy ◽  
Rényi Entropy ◽  
Image Descriptor ◽  
Local Image Descriptor ◽  
Local Image

Minimization of Error Functionals over Perceptron Networks

2008 ◽  
Vol 20 (1) ◽  
pp. 252-270 ◽  
Author(s):  
Věra Kůrková
Keyword(s):  
Gaussian Function ◽  
Regression Function ◽  
Rates Of Convergence ◽  
Oscillatory Behavior ◽  
Training Data ◽  
Partial Derivatives ◽  
High Regularity ◽  
Variational Norms ◽  
Derivatives Of ◽  
Error Dependence

Supervised learning of perceptron networks is investigated as an optimization problem. It is shown that both the theoretical and the empirical error functionals achieve minima over sets of functions computable by networks with a given number n of perceptrons. Upper bounds on rates of convergence of these minima with n increasing are derived. The bounds depend on a certain regularity of training data expressed in terms of variational norms of functions interpolating the data (in the case of the empirical error) and the regression function (in the case of the expected error). Dependence of this type of regularity on dimensionality and on magnitudes of partial derivatives is investigated. Conditions on the data, which guarantee that a good approximation of global minima of error functionals can be achieved using networks with a limited complexity, are derived. The conditions are in terms of oscillatory behavior of the data measured by the product of a function of the number of variables d, which is decreasing exponentially fast, and the maximum of the magnitudes of the squares of the L1-norms of the iterated partial derivatives of the order d of the regression function or some function, which interpolates the sample of the data. The results are illustrated by examples of data with small and high regularity constructed using Boolean functions and the gaussian function.

HSOG: A Novel Local Image Descriptor Based on Histograms of the Second-Order Gradients

2014 ◽  
Vol 23 (11) ◽  
pp. 4680-4695 ◽  
Author(s):  
Di Huang ◽  
Chao Zhu ◽  
Yunhong Wang ◽  
Liming Chen
Keyword(s):  
Second Order ◽  
Image Descriptor ◽  
Local Image Descriptor ◽  
Local Image

A LEAST‐SQUARES PROCEDURE FOR GRAVITY INTERPRETATION

Geophysics ◽  
1965 ◽  
Vol 30 (2) ◽  
pp. 228-233 ◽  
Author(s):  
Charles E. Corbató
Keyword(s):  
Least Squares ◽  
High Speed ◽  
Gravity Anomalies ◽  
The Body ◽  
Two Dimensional ◽  
Partial Derivatives ◽  
Dimensional Gravity ◽  
Gravity Interpretation ◽  
Relationship Of ◽  
Derivatives Of

A procedure suitable for use on high‐speed digital computers is presented for interpreting two‐dimensional gravity anomalies. In order to determine the shape of a disturbing mass with known density contrast, an initial model is assumed and gravity anomalies are calculated and compared with observed values at n points, where n is greater than the number of unknown variables (e.g. depths) of the model. Adjustments are then made to the model by a least‐squares approximation which uses the partial derivatives of the anomalies so that the residuals are reduced to a minimum. In comparison with other iterative techniques, convergence is very rapid. A convenient method to use for both the calculation of the anomalies and the adjustments is the two‐dimensional method of Talwani, Worzel, and Landisman, (1959) in which the outline of the body is polygonized and the anomalies and the partial derivatives of the anomaly with respect to the depth of a vertex on the body can be expressed as functions of the coordinates of the vertex. Not only depths but under certain circumstances regional gravity values may be evaluated; however, the relationship of the disturbing body to the gravity information may impose certain limitations on the application of the procedure.

Local image descriptor based on spectral embedding

2015 ◽  
Vol 9 (2) ◽  
pp. 278-289 ◽  
Author(s):  
Pu Yan ◽  
Jun Tang ◽  
Ming Zhu ◽  
Dong Liang
Keyword(s):  
Image Descriptor ◽  
Spectral Embedding ◽  
Local Image Descriptor ◽  
Local Image
Sign in / Sign up

Export Citation Format

Share Document