Unlock instant, AI-driven research and patent intelligence for your innovation.

Target detection model training method, model performance detection method and device

A target detection and model training technology, applied in the field of artificial intelligence, can solve the problem that the target detection algorithm cannot adaptively adjust the scale of the target area, and achieve the effect of improving the detection performance

Active Publication Date: 2021-12-17
西安电子科技大学广州研究院
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The invention provides a training method for a target detection model, a model performance detection method and a device to solve the technical problem that the existing target detection algorithm cannot adaptively adjust the scale of the target area

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
  • Target detection model training method, model performance detection method and device
  • Target detection model training method, model performance detection method and device
  • Target detection model training method, model performance detection method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0070] 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 creative efforts fall within the protection scope of the present invention.

[0071] see figure 1 , which is a schematic flowchart of a method for training a target detection model provided in Embodiment 1 of the present invention.

[0072] A target detection model training method provided by an embodiment of the present invention includes steps S11 to S14:

[0073] Step S11, clustering all target annotation frames on the first type of training pictures in the training picture set to obtain a plurality of initial regions; wherein, the train...

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 target detection model training method, and the method achieves the focusing of a region through a deep reinforcement learning network, generates an optimal region in which a target is easier to detect, generates a mixed training sample with a better training effect through the optimal region, and trains a target detection model through the mixed training sample. According to the invention, different target objects can be adaptively adjusted to a scale which is easy to learn by the target detection model for training, and the detection performance of the target detection model is improved. Correspondingly, the invention further provides a target detection model device and a target detection model performance detection method and device.

Description

technical field [0001] The invention relates to the technical field of artificial intelligence, in particular to a target detection model training method, a model performance detection method and a device. Background technique [0002] In the era of deep learning, object detection can be divided into two categories, one-stage object detection and two-stage object detection. [0003] One-stage object detection algorithms directly classify predefined anchor boxes and further refine and regress them through the detection head without generating proposal regions. In the single-stage target detection algorithm, EfficientDet designed a weighted bidirectional feature pyramid network to achieve simple and fast multi-scale feature fusion. EfficientDet also proposes a composite scaling method that uniformly scales resolution, depth, and width within a unified location and classification framework. [0004] Two-stage object detection algorithms usually first generate a set of region ...

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
IPC IPC(8): G06K9/62G06N3/08
CPCG06N3/08G06F18/23213Y02T10/40
Inventor 谢雪梅张裕博李旭阳石光明
Owner 西安电子科技大学广州研究院