Missile-borne image deblurring method based on generative adversarial network

A technology for deblurring and image, which is applied in biological neural network model, image enhancement, image analysis, etc. It can solve problems such as jitter, difficulty in accurate identification and blurring of missile-borne image targets, and improve detection accuracy and efficiency, and improve PSNR and SSIM indicators, the effect of eliminating ambiguous information

Pending Publication Date: 2022-01-18
BEIJING INSTITUTE OF TECHNOLOGYGY
View PDF0 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In order to solve the problem of serious blurring in the missile-borne image: the acquired image contains a lot of noise, jitter and blur characteristics, resulting in a serious decline in image quality, making it difficult to accurately identify the target in the missile-borne image, resulting in a decrease in the accuracy of image guidance

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Missile-borne image deblurring method based on generative adversarial network
  • Missile-borne image deblurring method based on generative adversarial network
  • Missile-borne image deblurring method based on generative adversarial network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0092] Embodiment: Deblurring the missile-borne image.

[0093] Such as figure 1 , 2 As shown, a method for deblurring missile-borne images based on an adversarial generation network disclosed in this embodiment includes the following steps:

[0094] Step 1: Make a training set of fuzzy data of missile-borne images.

[0095] Due to the difficulty in obtaining real strike scene data, according to the imaging characteristics of the missile-borne image, the scene of the field of view of the seeker is simulated by the scene generation server as the main source of image data, as shown in the attached image 3 As shown, the visual display screen generates the image visual of the seeker, and the development board collects the missile-borne image in real time through the camera, and the collected result is displayed on the monitor in real time, and this data is used as the main sample for training.

[0096] The specific implementation method of step 1 is:

[0097] Step 1.1: Manual...

Embodiment 2

[0154] Embodiment 2: Indirectly prove the effectiveness of the deblurring algorithm from the accuracy of target detection.

[0155] Step 1: Make a training set of missile-borne image fuzzy data.

[0156] Using the hardware-in-the-loop simulation test platform, 1000 pairs of missile-borne image fuzzy data sets with targets to be hit are produced, of which 900 pairs are used as training sets and 100 pairs are used as test sets.

[0157] Step 2: Establish a forward propagation model against the network.

[0158] The overall structure of the missile-borne image deblurring model is as follows: figure 2 As shown in (a), first input the blurred image into the generator and output an image to be judged, then take a corresponding clear image from the clear image set, and put the generator output and clear image into an image pair into the discriminator , to judge which of the two images is a clear image and which is a forged image by the generator. The goal of the generator is to g...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a missile-borne image deblurring method based on a generative adversarial network, and belongs to the technical field of missile-borne computer vision. According to the method, a fuzzy missile-borne image is deblurred by using a generative adversarial network, a deep convolutional generative adversarial network model comprising a generator and a discriminator is designed, the generator adopts a coding-decoding structure, a joint loss function is constructed, and continuous training is performed to generate a restored image of the missile-borne blurred image. A distinct image and an image forged by a generator are distinguished through a discriminator, the generator approaches the distinct image confusion discriminator, a network model reaches an expected index through adversarial training of two networks, and the network model after adversarial training is transplanted to a missile-borne computer for deblurring a missile-borne image and improving guidance precision. Besides, different fuzzy sources are simulated by establishing a vivid semi-physical simulation system for synthesizing motion blur, the problem that actual acquisition of missile-borne image data is difficult is solved, the training efficiency is improved, and the test cost is saved.

Description

technical field [0001] The invention belongs to the technical field of missile-borne computer vision, and in particular relates to a method for deblurring a missile-borne image. Background technique [0002] When the low-speed rolling projectile hits the target, due to the strapdown between the TV camera and the projectile body, the rapid movement, continuous rotation, and vibration of the projectile cause the quality of the acquired image to decline, resulting in distortion, blurring, and blurring of the collected field of view image. Problems such as defocusing will bring great difficulties to subsequent target detection and tracking, so the deblurring research of missile images is of great significance. [0003] Image deblurring refers to obtaining a clear image from a degraded image. It is an ill-posed problem and can be divided into two cases: blind image deblurring and non-blind image deblurring. Image non-blind deblurring refers to recovering a clear image when the b...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06T7/00G06T5/00G06K9/62G06N3/04G06N3/08G06V10/774G06V10/82
CPCG06T7/0002G06T5/003G06N3/08G06T2207/10004G06T2207/20081G06T2207/20084G06N3/045G06F18/214
Inventor 王少博张成苏迪冀瑞静陈志升
Owner BEIJING INSTITUTE OF TECHNOLOGYGY
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products