Image processing method and image processing device based on multiple tasks
An image processing, multi-task technology, applied in neural learning methods, instruments, biological neural network models, etc., can solve the problems of low target recognition accuracy, easy to ignore common features, and high difficulty in labeling
Pending Publication Date: 2022-05-27
SUN YAT SEN UNIV +1
View PDF0 Cites 1 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] However, due to the different characteristics of the target objects of each category, marking multiple categories of target objects in the same training image often requires the cooperation of multiple people or one person is familiar with multiple categories of target objects at the same time, which is difficult to label. Another recognition method based on an independent neural network tends to ignore the common features between multiple target objects, which leads to insufficient target recognition accuracy
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0025]Hereinafter, preferred embodiments of the present disclosure will be described in detail with reference to the drawings. In the following description, the same reference numerals are given to the same components, and repeated descriptions are omitted. In addition, the drawings are only schematic diagrams, and the ratio of dimensions between components, the shape of components, and the like may be different from the actual ones. It should be noted that the terms "comprising" and "having" and any variations thereof in the present disclosure, such as a process, method, system, product or device that includes or has a series of steps or units, are not necessarily limited to the clearly listed instead, may include or have other steps or elements not explicitly listed or inherent to the process, method, product or apparatus. All methods described in this disclosure can be performed in any suitable order unless otherwise indicated herein or otherwise clearly contradicted by co...
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
Login to View More Abstract
The invention discloses an image processing method and an image processing device based on multiple tasks. The image processing method comprises the following steps: acquiring an input image; the input image is input into a target recognition network for recognition to obtain a plurality of target outputs corresponding to a plurality of categories, each target output comprises the probability that each pixel point of the input image belongs to the corresponding category, and the target recognition network comprises a plurality of segmentation networks corresponding to a plurality of categories; each segmentation network takes an input image as an input and obtains each target output, the plurality of segmentation networks share an encoder, each segmentation network is provided with an independent self-attention mechanism module based on a self-attention mechanism and a decoder, and in each segmentation network, the self-attention mechanism module is arranged between the encoder and the decoder; and obtaining the target object of each category based on each target output. Therefore, the labeling difficulty can be reduced, and the recognition accuracy is relatively high.
Description
technical field [0001] The present disclosure generally relates to the field of artificial intelligence image processing, and specifically relates to a multi-task-based image processing method and image processing device. Background technique [0002] In recent years, artificial intelligence technology represented by deep learning has developed significantly, and the application of artificial intelligence in the field of image processing has also attracted more and more attention. In particular, deep learning-based neural networks have become effective means for object recognition. The neural network based on deep learning can identify the target object of interest from the image, and then can segment the target object to assist in the analysis of the target object. For example, the target object in the medical image can be identified through the trained neural network, and the target object in the natural image can also be identified. [0003] At present, for the recognit...
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
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06V20/10G06V10/26G06V10/774G06V10/764G06V10/82G06K9/62G06N3/04G06N3/08
CPCG06N3/04G06N3/08G06F18/214G06F18/24
Inventor 王学钦罗燕蒋宇康张可潘间英田婷
Owner SUN YAT SEN UNIV