Unlock instant, AI-driven research and patent intelligence for your innovation.
Image processing method, device, and electronic device
What is Al technical title?
Al technical title is built by PatSnap Al team. It summarizes the technical point description of the patent document.
An image processing, image technology, applied in the field of image processing
Active Publication Date: 2021-09-28
RICOH KK
View PDF2 Cites 0 Cited by
Summary
Abstract
Description
Claims
Application Information
AI Technical Summary
This helps you quickly interpret patents by identifying the three key elements:
Problems solved by technology
Method used
Benefits of technology
Problems solved by technology
Due to the unique perspective of wearable devices, the interaction process often faces unpredictable environmental changes, which brings more challenges to the recognition of hand images
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
Click on the blue label to locate the original text in one second.
Reading with bidirectional positioning of images and text.
Smart Image
Examples
Experimental program
Comparison scheme
Effect test
Embodiment 1
[0070] Embodiments of the present invention provide an image processing method, such as figure 1 shown, including:
[0071] Preliminary recognition step 101: using the appearance information of the image to be processed to generate a preliminary recognition result of the hand region;
[0072] Superpixel-level recognition step 102: divide the image into a plurality of superpixel regions whose area is smaller than a preset first threshold, and use the preliminary recognition results of the hand region to train a superpixel-level classifier, and pass the superpixel-level The classifier calculates the probability that each superpixel region is a hand region, and generates superpixel-level hand region recognition results;
[0073] Pixel-level recognition step 103: using the super-pixel-level hand region recognition result and the weight of the image to train a pixel-level classifier, using the pixel-level classifier to calculate the probability that each pixel of the image is a ha...
Embodiment 2
[0104] The embodiment of the present invention also provides an image processing device, such as Image 6 shown, including:
[0105] The preliminary recognition module 21 is used to generate a preliminary recognition result of the hand region by using the appearance information of the image to be processed;
[0106]The superpixel-level recognition module 22 is configured to divide the image into a plurality of superpixel regions whose area is smaller than a preset first threshold, and use the preliminary recognition result of the hand region to train a superpixel-level classifier, through the superpixel The pixel-level classifier calculates the probability that each superpixel region is a hand region, and generates superpixel-level hand region recognition results;
[0107] The pixel-level recognition module 23 is used to train a pixel-level classifier using the super-pixel-level hand region recognition result and the weight of the image, and calculate each pixel of the image ...
Embodiment 3
[0132] This embodiment provides an electronic device 30 for image processing, such as Figure 11 shown, including:
[0133] processor 32; and
[0134] memory 34 in which computer program instructions are stored,
[0135] Wherein, when the computer program instructions are executed by the processor, the processor 32 is made to perform the following steps:
[0136] A preliminary recognition step, using the appearance information of the image to be processed to generate a preliminary recognition result of the hand region;
[0137] The superpixel-level recognition step is to divide the image into a plurality of superpixel regions whose area is smaller than the preset first threshold, and use the preliminary recognition result of the hand region to train a superpixel-level classifier, and pass the superpixel-level classification The detector calculates the probability that each superpixel region is a hand region, and generates a superpixel-level hand region recognition result; ...
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 provides an image processing method, device, and electronic equipment. The image processing method includes: a preliminary recognition step, generating a preliminary recognition result of the hand region; a superpixel-level recognition step, dividing the image into multiple superpixel regions, and using the hand The superpixel-level classifier is trained on the preliminary recognition results of the hand region, and the probability of each superpixel region being a hand region is calculated by the superpixel-level classifier to generate the superpixel-level hand region recognition result; the pixel-level recognition step uses the superpixel-level Hand region recognition results and image weights train a pixel-level classifier, calculate the probability that each pixel of the image is a hand region through the pixel-level classifier, and generate a pixel-level hand region recognition result; the update step, according to the pixel-level hand region Update the preliminary recognition result of the hand region, the calculation parameters of the superpixel-level hand region recognition result, and the weight of the image, and return to the preliminary recognition step. The present invention can finely segment the hand image in the image.
Description
technical field [0001] The present invention relates to the technical field of image processing, in particular to an image processing method, device, and electronic equipment. Background technique [0002] In recent years, with the rapid development of augmented reality and virtual reality technology, the natural human-computer interaction technology of wearable devices has also received more and more attention. The user's hand is undoubtedly one of the most suitable media for augmented reality and virtual reality interaction. In natural human-computer interaction, the wearable device acquires the image of the user's hand, recognizes the position, posture and gesture of the user's hand, and then makes appropriate feedback to the user. The key step in this process is the recognition of the hand image. Due to the unique perspective of wearable devices, the interaction process often faces unpredictable environmental changes, which brings more challenges to the recognition of h...
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
Application Date:The date an application was filed.
Publication Date:The date a patent or application was officially published.
First Publication Date:The earliest publication date of a patent with the same application number.
Issue Date:Publication date of the patent grant document.
PCT Entry Date:The Entry date of PCT National Phase.
Estimated Expiry Date:The statutory expiry date of a patent right according to the Patent Law, and it is the longest term of protection that the patent right can achieve without the termination of the patent right due to other reasons(Term extension factor has been taken into account ).
Invalid Date:Actual expiry date is based on effective date or publication date of legal transaction data of invalid patent.
Login to View More 
Login to View More