scispace - formally typeset
Search or ask a question
Topic

Standard test image

About: Standard test image is a research topic. Over the lifetime, 5217 publications have been published within this topic receiving 98486 citations.


Papers
More filters
Patent
Yoshiaki Honda1
29 Sep 2011
TL;DR: In this paper, a first image data is stored in a first storage; second image data of a low resolution is generated by the first reducing; enlarged image data are generated; low resolution image data was generated from the first image through processing that is different from that performed by first reducing, the low resolution images had the same pixel count as the first images and a lower resolution than that of the first data.
Abstract: In an image capturing apparatus, a first image data is stored in a first storage; second image data of a low resolution is generated by the first reducing; enlarged image data is generated; low resolution image data is generated from the first image data through processing that is different from that performed by the first reducing, the low resolution image data have the same pixel count as the first image data and a lower resolution than that of the first image data; one of first image processing in which the first image data is combined with the enlarged image data or second image processing in which the first image data is combined with the low resolution image data is executed; and the processing is switched between the first image processing and the second image processing, according to a shooting operation.

24 citations

Patent
27 Feb 2013
TL;DR: In this article, a no-reference image quality evaluation method based on sparse representation in natural scene statistics is proposed, which mainly solves the problem that objective evaluation score and subjective evaluation score are different in the prior art that the original image is unknown.
Abstract: The invention discloses a no-reference image quality evaluation method based on sparse representation in natural scene statistics, and mainly solves the problem that objective evaluation score and subjective evaluation score are different in the prior art that the original image is unknown. The method comprises the steps of decomposing a sub-band for the image by wavelet transform and extracting effective characteristics of the image according to natural scene statistic characteristics; extracting the natural scene statistic characteristics to establish a characteristic dictionary by combining a series of images with different contents and distortion types; in the characteristic dictionary established, testing the natural scene statistic characteristics of the image by sparse representation; and weighing different average subjective score value DMOS of associated image by using a spare representation coefficient linear to finally obtain the evaluation value of the test image quality. The mtehod is suitable for various distortion types and the subjective and objective evaluations are consistent and good, and can be used to evaluate the effectiveness to an image processing method.

24 citations

Patent
Hamanaka Masahiko1
08 Jul 2003
TL;DR: In this paper, the authors propose a method to search a reference image stored in a database from an input image of an object imaged with a different pose and a different illumination condition.
Abstract: Even when a small number of reference images are available for each object, it is possible to search at high speed a reference image stored in a database from an input image of an object imaged with a different pose and a different illumination condition. A reference image matching result storage section (50) inputs reference images from a reference image storage section (70) and stores in advance results of matching of the input images with representative 3-dimensional object models of a representative 3-dimensional object model storage section (20). According to each representative 3-dimensional object model, image generation means (30) generates a comparison image having an input condition similar to the input image obtained from the image input means (10). Image matching means (40) calculates similarity between the input image and the image generated. Result matching means (60) calculates similarity between the matching result of the image matching means (40) and the reference image stored in the reference image matching result storage section (50), extracts reference images having similar matching results in the descending order of the similarity, and displays them on result display means (80).

24 citations

Journal ArticleDOI
TL;DR: A novel blur metric based on Multiscale SVD fusion (M-SVD) fuses different sub-bands of the selected singular values (SVs) in multiscale image windows, which could drastically reduce the chances of false positives for blur detection and overcome the difficulty that the sharp region is misjudged for a blur region because of its smooth texture.

24 citations

Book ChapterDOI
13 Apr 1996
TL;DR: An experimental investigation of the recognition performance of two approaches to the representation of objects for recognition by constructing an eigenvector space to compute efficiently the distance between a new image and any image in the database.
Abstract: This paper describes an experimental investigation of the recognition performance of two approaches to the representation of objects for recognition. The first representation, generally known as appearance modelling, describes an object by a set of images. The image set is acquired for a range of views and illumination conditions which are expected to be encountered in subsequent recognition. This image database provides a description of the object. Recognition is carried out by constructing an eigenvector space to compute efficiently the distance between a new image and any image in the database. The second representation is a geometric description based on the projected boundary of an object. General object classes such as planar objects, surfaces of revolution and repeated structures support the construction of invariant descriptions and invariant index functions for recognition.

24 citations


Network Information
Related Topics (5)
Feature extraction
111.8K papers, 2.1M citations
91% related
Image segmentation
79.6K papers, 1.8M citations
91% related
Image processing
229.9K papers, 3.5M citations
90% related
Convolutional neural network
74.7K papers, 2M citations
90% related
Support vector machine
73.6K papers, 1.7M citations
90% related
Performance
Metrics
No. of papers in the topic in previous years
YearPapers
20231
20228
2021130
2020232
2019321
2018293