Object detection system based on darknet framework
An object detection and framework technology, applied in the field of detection systems, can solve the problems of difficult training, complex process, slow speed, etc., and achieve the effect of high accuracy, simple training and accurate positioning
Inactive Publication Date: 2018-01-30
SHENZHEN SANBOT INNOVATION INTELLIGENT CO LTD
View PDF3 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
This type of method has a complicated process and has the problems of slow speed and difficult training.
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
[0016] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.
[0017] see Figure 1-3 , an object detection system based on the darknet framework, which includes a preprocessing unit, a model training unit, and a location recognition unit.
[0018] Preprocessing unit: This step mainly includes collecting a suitable picture set, labeling each picture in the picture set, and then converting the labeling information into the labeling format required for darknet training, and generating corresponding training according to a c...
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 object detection system based on a darknet framework. The object detection system comprises a preprocessing unit, a model training unit and a positioning and identificationunit. The object detection system is simple in model training, the whole system is end-to-end (inputting a picture, outputting a detection frame), and real-time detection can be realized. The object detection system adopts full-map information to make predictions. Unlike a sliding window method and a region proposal-based method, a network model can utilize the full-map information during the training and prediction process, and the positioning of a target frame is more precise. The object detection system can learn the general information of a target, and has certain universality. The objectdetection system adopts natural images for training the network model and adopts artistic image for prediction, has much higher accuracy rate than other target detection methods. The object detectionsystem can be adapted to different application scenarios and detected objects, and just needs to be trained by using corresponding data sets.
Description
technical field [0001] The invention relates to a detection system, in particular to an object detection system based on a darknet framework. Background technique [0002] Object recognition is a research direction in computer vision, and it is also a popular research field at present. Today, as people's needs continue to grow, object recognition is playing a pivotal role in safety, technology, and economy. The security field and traffic supervision departments have also put forward urgent requirements for object recognition, so the study of object recognition is of great significance to the future of society. very important meaning. [0003] The traditional target detection system adopts the deformable parts models (DPM) method, proposes the target area through the sliding frame method, and then uses the classifier to realize the recognition. Recent R-CNN-like methods employ region proposal methods that first generate latent bounding boxes and then employ a classifier to ...
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): G06K9/32G06K9/34G06K9/62
Inventor 周海明林绿德庄永军
Owner SHENZHEN SANBOT INNOVATION INTELLIGENT CO LTD