Unlock instant, AI-driven research and patent intelligence for your innovation.

2D image salient target detection method and system

A target detection and salience technology, applied in the direction of neural learning methods, instruments, biological neural network models, etc., can solve the problems of false detection in background areas, time-consuming training, and real-time detection needs to be improved, so as to meet the accuracy and real-time performance Requirements, solutions to incomplete testing, and the effect of ensuring the accuracy of testing

Active Publication Date: 2021-03-05
SHANDONG UNIV
View PDF7 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The inventors found that the current 2D image salient target detection faces challenges such as blurred edges of the detection area, incomplete detection of salient targets, and false detection of background areas, which affects the detection accuracy. In addition, the related convolutional network model achieves high precision. It takes time to train the network in the early stage, and the real-time detection in the later stage needs to be improved

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
  • 2D image salient target detection method and system
  • 2D image salient target detection method and system
  • 2D image salient target detection method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0033] This embodiment discloses a method for detecting salient objects in 2D images, which is applied to practical fields such as unmanned driving, underwater archaeology, and three-dimensional scene reconstruction.

[0034] Combine below figure 1 Take the images collected by unmanned driving as an example to explain in detail:

[0035] The salient target detection method of this embodiment includes:

[0036] Use the dual-pool U-shaped network to detect the salient objects in each frame of the video of the driverless driving scene, remove a large amount of redundant information collected in the video image, and prepare for the perception, planning and decision-making of the next stage of driverless driving .

[0037] Among them, the training process of the dual-pool U-shaped network is:

[0038] The images in the training set are firstly subtracted from the mean, then randomly flipped horizontally for data enhancement, and finally used as network input to train the dual-po...

Embodiment 3

[0110] The purpose of this embodiment is to provide a computing device, including a memory, a processor, and a computer program stored in the memory and operable on the processor. When the processor executes the program, the method in the first embodiment above is implemented. specific steps.

Embodiment 4

[0112] The purpose of this embodiment is to provide a computer-readable storage medium.

[0113] A computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the specific steps of the method in the first embodiment above are executed.

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 provides a 2D image salient target detection method and system, and the method comprises the steps: obtaining a to-be-detected video, and extracting 2D images in the same scene from thevideo; performing salient target detection on each frame of image in the 2D image in the unmanned scene video by using a double-pooling U-shaped network; wherein during salient target detection, the double-pooling U-shaped network is configured to firstly perform preprocessing, channel conversion and feature refinement on each frame of image in an unmanned scene video to obtain a refined feature map, and then perform bilinear layer-by-layer up-sampling and convolution operation to obtain a multi-scale prediction feature map; obtaining a multi-scale prediction graph through prediction convolution, and obtaining a salient map through the connection and convolution operation of the edge output prediction graph of each level. Salient target detection is performed on each frame of image in theunmanned scene video by using the double-pooling channel network so that the image processing speed can be enhanced while the detection accuracy can be guaranteed.

Description

technical field [0001] The disclosure belongs to the technical field of image salient object detection, and in particular relates to a method and system for detecting a 2D image salient object. Background technique [0002] The statements in this section merely provide background information related to the present disclosure and do not necessarily constitute prior art. [0003] The essence of image salient target detection is to perform binary classification of salient or non-salient prediction on each pixel in the picture, so as to obtain a prediction map with salient target information. The scene information contained in the image is often numerous and complex, and thanks to the human eye's visual attention mechanism, humans can always extract the most significant and critical information from the image for processing, so as to make fast and accurate decisions. Image salient target detection is based on the study of human visual attention mechanism, using computer to simu...

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): G06K9/00G06N3/04G06N3/08
CPCG06N3/08G06V20/41G06V20/56G06V2201/07G06N3/045Y02T10/40
Inventor 陈振学孙露娜刘成云闫星合
Owner SHANDONG UNIV