KEYWORDS: Feature extraction, Structured light, 3D modeling, Image processing, 3D acquisition, Environmental sensing, 3D vision, Visual process modeling, Inspection, 3D image processing
Robot 3D vision inspection plays a very important role in intelligent manufacturing process such as automated picking, obstacle avoidance, path planning and so on. Recently, there is a need for a fast detection method that has applicability to complex environments, strong anti-interference capabilities, and balances speed and accuracy to meet the above requirements. An environment feature detection method based on laser-assisted machine vision is proposed. By illuminating the grid structure to the target scene, the binocular camera is used to collect the grid image on the surface of the target scene. Then A two-step feature extraction method is proposed, which is locating the feature position quickly first, and then accurately obtaining the coordinate of the feature point. Firstly, an improved fast extraction method is proposed to realize the fast recognition of feature points. Secondly, in the aspect of accurate acquisition, a new improved steger fitting method is proposed to accurately extract the position of feature points. Finally, fast matching and reconstruction of the exact position of feature points on the two images collected by binocular camera are implemented to achieve fast and high precision 3D detection. This experiment has verified the rationality of the system scheme, the correctness, the precision and effectiveness of the relevant methods.
In order to solve the problem of slow detection efficiency and low accuracy in continuous detection of metal prints. A
detection method for printing defects and surface damage of metal prints based on super-resolution reconstruction is
proposed. The core content of this method is the use of image quadratic linear interpolation. First, according to the
spectral reflectance characteristics of different surfaces of metal prints, a method of imaging features that can clearly
reflect the image is proposed. In this paper, an image super-resolution registration method based on quadratic linear
interpolation is proposed, and finally a method based on image threshold segmentation to reduce noise interference is
proposed, which effectively improves the detection accuracy. Experiments show that the error of the detection method
for printing defects and surface damage of metal prints based on super-resolution reconstruction is less than 0.03mm.
Field experiment results show that this method is an effective non-destructive testing method for metal printing.
Accurate acquisition of image features is particularly important for environmental perception of robots. It is difficult to obtain the image features accurately and effectively because of the interference of high-light reflection of the parts’ surface. To solve this problem an improved distortion feature information restoration method is proposed in this paper. First, an image acquisition method is proposed, which can successfully reduce the effect of high-light reflection. Then, an image quality evaluation method based on cosine similarity is proposed to realize the accurate evaluation image quality of highlight reflection image. Finally, an image quality compensation method is proposed to effectively reduce the effect of high light distortion on feature extraction. Contrastive experiments fully prove that the method proposed in this paper can effectively improve the problem of feature recognition caused by local strong light interference, and the accuracy of image surface feature extraction after compensation is significantly improved.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.