Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Deep learning-based redundant object visual detection system and method in mechanical assembly

A deep learning and visual detection technology, applied in the field of artificial intelligence, can solve problems such as affecting the normal operation of high-reliability equipment, large human factors, safety accidents, etc., and achieve the effect of meeting real-time detection requirements, high recognition accuracy, and strong applicability

Active Publication Date: 2019-10-29
ZHEJIANG UNIV
View PDF6 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

If these redundant substances are left in the equipment, it will leave serious safety hazards, which may affect the normal operation of high-reliability equipment, and even cause malfunctions and safety accidents.
[0003] After long-term development, the current redundant detection and control methods mainly include visual and auditory detection, endoscopic detection, X-ray perspective detection, ultrasonic detection, Matera detection and particle collision noise detection, etc. , with more and stricter detection steps
However, most automatic detection methods are only suitable for fully assembled objects, and manual inspection is often used in assembly, and there are still problems such as large human factors and easy missed inspections.

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
  • Deep learning-based redundant object visual detection system and method in mechanical assembly
  • Deep learning-based redundant object visual detection system and method in mechanical assembly
  • Deep learning-based redundant object visual detection system and method in mechanical assembly

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment

[0046] The method for visual detection of redundant objects in mechanical assembly includes the following steps:

[0047] Step 1: Turn on the light source and camera. Multiple cameras collect image signals of the area where the assembly is located during mechanical assembly from multiple angles, and convert the image signals into digital electrical signals and send them to the vision controller.

[0048] Step 2: The vision controller is connected to the camera through the interface line to collect a large number of images of the historical assembly process. After preprocessing, the objects in the images are manually identified, and the type number, center coordinates, width and height of each object are marked, and the data is constructed. The data set is randomly divided into training set and validation set according to the ratio of 4:1.

[0049] The type number of the object is marked according to the type number in the object list. The object list contains all the types of ...

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 deep learning-based redundant object visual detection system and method in mechanical assembly. The detection system involves a worktable, an angle-adjustable camera, a lightsource, a visual controller and the like, wherein by processing and analyzing images acquired from the camera, the visual controller is used for identifying objects in the images and then distinguishing assembly parts and redundant objects. The detection method comprises the following steps of acquiring the images in a multi-angle assembly area, preprocessing the images, inputting the images intoa trained target detection network model for feature extraction to predict the positions and the types of the objects, then judging whether the objects belong to redundant objects, and marking the positions of the redundant objects and giving an alarm. The system and method can be used for assisting people with real time and multi-angle detection on the redundant objects in the specific area in the assembling process; the system and the method have the advantages of being high in detection accuracy, high in real-time performance, flexible in use and the like; and the introduction of the redundant objects in the assembling process can be reduced, and the product reliability is enhanced.

Description

technical field [0001] The invention belongs to the field of artificial intelligence, and in particular relates to a system and method for visual detection of redundant objects based on deep learning in mechanical assembly. Background technique [0002] Residues refer to all substances in the product that are entered from the outside or generated inside and have nothing to do with the specified state of the product. In the assembly process of large-scale equipment with high reliability and high safety, due to the complex structure of equipment and various production processes and assembly procedures, it is very easy to introduce redundant materials. For example, improper operation of workers may bring in objects such as screws, washers, hair strands, and rag residues; processes such as welding and machining may introduce redundant objects such as welding slag and metal chips. If these redundant substances are left in the equipment, it will leave serious safety hazards, whic...

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): B07C5/342B07C5/36B07C5/38G06K9/00G06K9/62G06N3/04G06N3/08
CPCB07C5/3422B07C5/361B07C5/38G06N3/08G06V20/10G06V2201/07G06V2201/12G06N3/045G06F18/23213G06F18/214
Inventor 王宣银汤继祥林天培
Owner ZHEJIANG UNIV
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
Eureka Blog
Learn More
PatSnap group products