A method for making industrial robot visual semantic segmentation database
An industrial robot, semantic segmentation technology, applied in image data processing, still image data indexing, still image data retrieval, etc. The effect of identifying tasks, realizing artificial intelligence, and simple and convenient methods
Active Publication Date: 2020-09-25
KUNMING UNIV OF SCI & TECH
View PDF4 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Standard databases for semantic segmentation that are truly suitable for visual inspection and recognition tasks on industrial robot platforms are extremely rare, which restricts the application and expansion of deep learning in the field of machinery industry
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 1
[0034] Embodiment 1: as Figure 1-4 As shown, a method for making an industrial robot visual semantic segmentation database. First, use image acquisition equipment to collect several images of five common mechanical parts and tools (bearings, screwdrivers, gears, pliers, and wrenches) with different backgrounds and different samples. ; Then utilize PhotoShop software to process the target area and the background matting of each category that appear in the image, utilize a color to fill the target area and the background area respectively, and carry out stroke operation to the target area; utilize Matlab software Then classify the pixel values of the background area and the five categories of target areas in the image as 0 to 5, convert the image into cdata format, and superimpose the cdata image with the colormap to complete the ground-truth production of the original image.
[0035] The concrete steps of described method are as follows:
[0036] Step1. First, use image acq...
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 relates to an industrial robot visual sense meaning segmentation database manufacturing method, and belongs to the digital image processing technical field; the method comprises the following steps: collecting a plurality of images of machine parts and tools under different backgrounds and samples; cutting out target areas and backgrounds of all types in the images, refilling, and outlining the target areas; classifying pixel values of background areas and class areas in the images, converting the images into a cdata format, stacking the cdata images with a colormap, and finishing original image Ground-truth fabrication. The formed database can be applied to depth learning network model training, verification and detection, thus enabling the depth learning network to detect and identify normal machine parts and tools; in addition, the method can build databases of many types according to real demands, thus satisfying multi-type target detection and tasks.
Description
technical field [0001] The invention relates to a method for making an industrial robot visual semantic segmentation database, belonging to the technical field of digital image processing. Background technique [0002] In recent decades, with the rapid popularization of Industry 4.0, the detection and recognition of targets by industrial robots has become one of the research hotspots in the field of machine vision, and has become an indispensable part of many automated and intelligent manufacturing production systems. The detection and identification of mechanical parts and tools came into being along with modern industrial production. In modern intelligent manufacturing, especially in automated assembly work, it is a very critical work link to be able to detect and identify parts and tools under different backgrounds. It is of great significance in improving the efficiency of inspection, assembly, sorting, and storage in manufacturing work, and reducing labor costs. [000...
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 Patents(China)
IPC IPC(8): G06T7/11G06T7/194G06F16/51
CPCG06F16/51G06T7/11G06T7/194G06T2207/10024G06T2207/30164
Inventor 张印辉张春全何自芬武玉琪张越
Owner KUNMING UNIV OF SCI & TECH