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
25 Jul 2012
TL;DR: In this paper, a two-dimensional digital dynamic image platform is proposed, which is a linear vertical platform characterized in that: setting continuous images with action information based on the length and width in the running direction of a train within a range of 28 meters from the outside of a railway line safety protection area on both sides of a linear train track, and changing the image conversion speed by setting a state of a running track and the length of each image.
Abstract: The invention discloses a two-dimensional digital dynamic image platform, which is a linear vertical platform characterized in that: setting continuous images with action information based on the length and the width in the running direction of a train within a range of 28 meters from the outside of a railway line safety protection area on both sides of a linear train track, and changing the image conversion speed by setting a state of a running track and the length of each image. A linear dynamic image is placed on the platform, when siting in a fast-moving train, people sees that the speed of a thing outside is also high through a window, when a line of sight of a person rests on the image on the platform, the next image comes, before one image does not disappear, or when the vision temporarily stays in a brain of the person, the dynamic image is transferred to the person, a smooth visual change effect is achieved, an image of advertised enterprise products is made alive, and people in journey can remember the products of an enterprise more clearly.
Patent
03 Jun 2021
TL;DR: In this article, an obstacle recognition device capable of verifying a camera posture parameter to be used for image conversion and capable of defining a relationship between a device camera and the current environment, thereby achieving improved obstacle recognition reliability by simultaneously improving both the accuracy of obstacle detection and accuracy in calculating the distance to the detected object.
Abstract: Provided is an obstacle recognition device capable of verifying a camera posture parameter to be used for image conversion and capable of defining a relationship between a device camera and the current environment, thereby achieving improved obstacle recognition reliability by simultaneously improving both the accuracy of obstacle detection and accuracy in calculating the distance to the detected object even if there is a change in environmental conditions in which a vehicle is traveling. This invention comprises: acquiring a first difference (∆A) relating to the posture of a camera based on road surface shapes at different time instants; acquiring a second difference (∆R) relating to the amount of movement of characteristic points within images captured at said different time instants; and verifying, by employing the first difference (∆A) and the second difference (∆R), a camera posture parameter to be used for image conversion for calculating a differential image.
Patent
24 May 2018
TL;DR: In this paper, a communication system consisting of a portable terminal 14 and a control device 12 of an industrial machine that molds a molded product, and performs mutual communication between the portable terminal and the control device is presented.
Abstract: PROBLEM TO BE SOLVED: To provide a communication system and a portable terminal capable of automatically setting image data picked up by the portable terminal so that the image data can be used on a control device side of an industrial machine.SOLUTION: A communication system 10 includes a portable terminal 14 and a control device 12 of an industrial machine that molds a molded product, and performs mutual communication between the portable terminal 14 and the control device 12. The portable terminal 14 includes: an imaging unit 44 that picks up an imaging object; an image conversion unit 58 that converts picked-up image data picked up by the imaging unit 44 in accordance with a format for an image corresponding to the control device 12 to generate the converted image data; a setting data creation unit 60 that creates setting data of the converted image data so that the converted image data can be used on the side of the control device 12; and a data transmission/reception unit 46 that performs communication with the control device 12 and transmits the converted image data and the setting data to the control device 12.SELECTED DRAWING: Figure 1
Patent
30 Nov 2018
TL;DR: In this paper, a real-time binocular stylization rendering method based on deep learning is proposed, where an image conversion network is trained for a fixed style at first, and when running is carried out, the network is decomposed into a coder part and a decoder part; when rendering is performed, information, such as an original image, a parallax image and a motion field, is rendered through a three-dimensional scene; and furthermore, stylization is performed through post-processing based on the image conversion networks, so that a complete binocular animation is
Abstract: The invention discloses a real-time binocular stylization rendering method based on deep learning. The method is as follows: an image conversion network is trained for a fixed style at first; when running is carried out, the network is decomposed into a coder part and a decoder part; when rendering is carried out, information, such as an original image, a parallax image and a motion field, is rendered through a three-dimensional scene; and furthermore, stylization is carried out through post-processing based on the image conversion network, so that a complete binocular animation is rendered. Compared with the individual-frame direct stylization method, visual flashing and defects can be greatly reduced through the method in the invention; and, compared with the optimization-based method, the efficiency of the method in the invention is increased by two orders of magnitude.
Patent
31 Aug 2017
TL;DR: In this article, a character recognition section 102 extracts first text data which are recognized by performing character recognition on image data, an image conversion section 103 generates converted image data by changing a resolution of the image data.
Abstract: PROBLEM TO BE SOLVED: To provide an image processing apparatus capable of reducing a data amount of image data including characters without imposing a work load on a user, an image processing method and an image processing program.SOLUTION: A character recognition section 102 extracts first text data which are recognized by performing character recognition on image data, an image conversion section 103 generates converted image data by changing a resolution of the image data, and the character recognition section 102 extracts second text data which are recognized by performing character recognition on the converted image data. A replacement judgement section 104 compares the first text data to the second text data and judges whether a degree of mismatching indicating a rate of mismatching the second text data with the first text data is settled within a predetermined range. If it is judged that the degree of mismatching is settled within the predetermined range, an image replacement section 105 replaces the image data with the converted image data.SELECTED DRAWING: Figure 1

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