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 published on a yearly basis
Papers
More filters
Patent•
28 Mar 1997
TL;DR: In this article, the problem of image transmission efficiency by dividing an original image so as to have these divided images similar to each other, transmitting the divided images after encoding them and giving interpolation to the wrong group via the corresponding part of another similar divided image if an error is detected in a received image is solved.
Abstract: PROBLEM TO BE SOLVED: To improve the image transmission efficiency by dividing an original image so as to have these divided images similar to each other, transmitting the divided images after encoding them and giving interpolation to the wrong group via the corresponding part of another similar divided image if an error is detected in a received image. SOLUTION: An image conversion part 10 rearranges an original image to be inputted into plural images of high correlation on every pixel line. Then an encoding part 2 compresses and encodes the converted image data by an H. 261 encoding system or an MPEG1 system and builds the encoded image data into a transmission packet in every encoding unit GOB(group of block) to sends them to a transmission line 4 with addition of an error detection code. When an error is detected at the receiving side via an error detection part 12, an error interpolation part 13 performs the interpolation of the error by copying a normally received corresponding part from a divided image other than that having a wrong part. Then an image inverse conversion part 11 reconfigurates the divided images and outputs them. Thus a transmission error is easily interpolated with high precision and the accuracy of the reconfigurated images can be improved.
5 citations
Patent•
20 Oct 2011
TL;DR: In this paper, a vehicle circumference monitoring device is configured to perform image recognition to a captured image by an in-vehicle camera of a vehicle and detect an obstacle around the vehicle based on the consistency of shape between the captured image and a reference image obtained by imaging an obstacle by a different reference camera.
Abstract: PROBLEM TO BE SOLVED: To provide a vehicle circumference monitoring device, capable of achieving high-accuracy obstacle detection at low cost.SOLUTION: The vehicle circumference monitoring device 100 is configured to perform image recognition to a captured image by an in-vehicle camera of a vehicle and detect an obstacle around the vehicle based on the consistency of shape between the captured image and a reference image obtained by imaging an obstacle by a different reference camera. The device includes an image conversion unit 140 which converts the captured image to an image approximated to a vision as viewed from a position corresponding to the point of view of the reference camera; and an image recognition unit 150 which performs image recognition to the converted captured image based on the reference image to detect the obstacle.
5 citations
Patent•
20 Oct 2017
TL;DR: In this paper, an image conversion network processing method is executed based on a trained first network and comprises the following steps: obtaining a first image uploaded by the terminal; inputting the first image into the first network to obtain a second network corresponding to the style of the first images; transmitting the second network to the terminal, so that the terminal performs style processing on a to-be-processed second image by using the second networks.
Abstract: The invention discloses an image conversion network processing method, a server, a terminal, a computing device and a storage medium. The image conversion network processing method is executed based on a trained first network and comprises the following steps: obtaining a first image uploaded by the terminal; inputting the first image into the first network to obtain a second network corresponding to the style of the first image; transmitting the second network to the terminal, so that the terminal performs style processing on a to-be-processed second image by using the second network. By adoption of the technical scheme, the corresponding image conversion network can be obtained quickly by the trained first network, thereby improving the image conversion network obtaining efficiency and optimizing the image conversion network processing mode.
5 citations
Patent•
29 May 2002
TL;DR: In this paper, the authors present an image communication apparatus, an image processing apparatus, a communication method, an imaging processing method, and a program, which is controlled to transmit a conversion result image without performing any image conversion process to the former image.
Abstract: The present invention relates to an image communication apparatus, an image processing apparatus, an image communication method, an image processing method, and a program. According to the present invention, in a case where a conversion result image corresponding to a former image has been stored in a memory, it is controlled to transmit this conversion result image without performing any image conversion process to the former image.
5 citations
Patent•
22 Jan 2009
TL;DR: In this paper, a low pass filter based on a prevention band frequency was used to reduce the sample number of the image data by 1/M, which resulted in the high quality compressed data image and reproduced image that do not greatly damage the original signal harmonic component.
Abstract: PROBLEM TO BE SOLVED: To provide a device for carrying out data conversion and reproduction to prevent the image quality of moving image data from being deteriorated, and a method. SOLUTION: In carrying out downsampling for reducing the sample number of the image data by 1/M, a band restriction is conducted to the moving image data by a low pass filter based on a prevention band frequency fs whose harmonic component of a degree of K or higher out of harmonic components generated by down sampling is not superimposed with an original signal component. This results in generating compressed data by carrying out the downsampling to the data to which the band restriction is conducted. This configuration not only prevents folding distortion generated by a high degree harmonic but also generates the high quality compressed data image and reproduced image that do not greatly damage the original signal harmonic component. COPYRIGHT: (C)2009,JPO&INPIT
5 citations