scispace - formally typeset
Search or ask a question
Topic

Image conversion

About: Image conversion is a research topic. Over the lifetime, 2490 publications have been published within this topic receiving 19077 citations.


Papers
More filters
Patent
20 Aug 2013
TL;DR: In this article, a depth map compensation device, a method thereof, a stereoscopic image conversion device using the same, and a method of compensating boundary properties not coincided with a source image as much as possible with consideration with the source image is provided.
Abstract: PURPOSE: A depth map compensation device, a method thereof, a stereoscopic image conversion device using the same, and a method thereof are provided to compensate boundary properties of a depth map not coincided with a source image as much as possible with consideration with the source image. CONSTITUTION: A filtering unit (210) performs sharp characteristic improvement filtering or noise filtering in respect to a two-dimensional input image. A region setting unit (220) selects a boundary surface in which a change in properties of a pixel is largest from the filtered image and dividing the selected boundary surface into a compensation region, a neighboring region, and an edge region depending on the extent of the compensation. A depth value compensation unit (230) performs interpolation using the filtered image and the depth map formed in advance in respect to the compensation region and generates the depth map compensated through the compensation of depth values of the neighboring region and the edge region. [Reference numerals] (210) Filtering unit; (220) Region setting unit; (230) Depth value correction unit; (AA) Circle image; (BB) Depth map; (CC) Corrected depth map

3 citations

Patent
06 Apr 1991
TL;DR: In this paper, the authors proposed a read-out method for reading out a radioactive ray image without deteriorating sharpness or sensitivity and provided a panel which can be used favourbly in the readout method.
Abstract: PURPOSE: To obtain a read-out method for reading out a radioactive ray image without deteriorating sharpness or sensitivity and a radioactive ray image conversion panel which can be used favourbly in the read-out method. CONSTITUTION: In the reading-out method for a radioactive ray image, the radioactive ray image information is stored by radiating the radioactive ray which transmits through a photographed body on a radioactive ray image conversion panel having at least a brilliant fluorescent body layer 3 and a low refractive index layer 2 formed on the second protecting layer 1 side between the first and second protecting layers 4 and 1. The radioactive ray image information is read out through the excitation by the brilliant exciting light from the first protecting layer 4 side of a radioactive ray image conversion panel. Further, the second protecting layer 1 on the radioactive ray image conversion panel has a light shielding layer. COPYRIGHT: (C)1992,JPO&Japio

3 citations

Patent
19 Feb 2019
TL;DR: In this article, the authors reveal image processing methods and devices and devices, equipment and a memory medium and belong to the field of the Internet, which comprises the steps of inputting an original image into an ad-ecoding network according to an image conversion instruction, and outputting a first feature image of the original image; inputting the first image into a plurality of conversion networks corresponding to at least one piece of conversion demand information in sequence, and then outputting second feature images; and inputting second image images into a reestablishment network, and finally outputting the second
Abstract: The invention discloses image processing methods and devices, equipment and a memory medium and belongs to the field of the Internet. A method comprises the steps of inputting an original image into adecoding network according to an image conversion instruction, and outputting a first feature image of the original image; inputting the first feature image into a plurality of conversion networks corresponding to at least one piece of conversion demand information in sequence, and outputting second feature images; and inputting the second feature images into a reestablishment network, and outputting a target image. According to the methods, the devices, the equipment and the memory medium, when a plurality of features are converted, only single-time decoding and single-time reestablishment need to be carried out, so when many features exist, an image processing process becomes concise and smooth.

2 citations

Patent
31 Jan 2002
TL;DR: In this paper, the registration system of pre- registration or post-registration can be appropriately set and changed in accordance with the installation form of a reader or console, and an image reading method.
Abstract: PROBLEM TO BE SOLVED: To provide an image reader in which the registration system of pre- registration or post-registration can be appropriately set and changed in accordance with the installation form of a reader or console, and an image reading method. SOLUTION: This image reader is composed of a reader 2 for reading image data from a radiographic image conversion plate with radiographic images recorded thereon as latent images, a controller 3 for controlling reading by the reader by inputting the read condition of the radiographic image conversion plate and a job manager 4 provided with a means 23 for storing the correspondence of the image reader and the registration system and storing the correspondence of an ID and the read condition in the pre-registration or post-registration and a search means 24 for performing a search by using a prescribed key. The image reader can meet any registration system by presetting the registration systems of the respective devices, searching a photographing condition with a plate ID as a key in the pre-registration and searching the photographing condition from the correspondence of an ID dedicated to the post-registration, the input order of the photographing condition and the inserting order of cassettes in the post-registration. COPYRIGHT: (C)2003,JPO

2 citations

Patent
24 May 2005
TL;DR: An image processing system includes an information processing apparatus for editing edit text data for processing an image and converting the edit text into a command for processing the image, and an image processing apparatus to process the image on the basis of the command as discussed by the authors.
Abstract: An image processing system includes an information processing apparatus for editing edit text data for processing an image and converting the edit text data into a command for processing the image, and an image processing apparatus for processing the image on the basis of the command. The information processing apparatus includes an editing section, a command conversion section, a processing section, an image conversion section, and an output section. The image processing apparatus includes an obtaining section, a recognition section, and a processing section.

2 citations


Network Information
Related Topics (5)
Image processing
229.9K papers, 3.5M citations
84% related
Feature (computer vision)
128.2K papers, 1.7M citations
83% related
Pixel
136.5K papers, 1.5M citations
82% related
Feature extraction
111.8K papers, 2.1M citations
80% related
Image segmentation
79.6K papers, 1.8M citations
79% related
Performance
Metrics
No. of papers in the topic in previous years
YearPapers
202132
202074
2019117
2018115
2017100
2016107