Vehicle target detection method and device and computer readable storage medium
A target detection and vehicle detection technology, applied in the field of target detection, can solve the problem that the target detection model cannot take into account the detection accuracy and detection speed, and achieve the effect of enhancing robustness, improving detection accuracy, and reducing the amount of calculation
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
no. 1 example
[0033] With the rise of parallel computing paradigms represented by graphics processing units (GPUs), the advantages of convolutional neural networks in image feature extraction have gradually emerged. In the ImageNet image classification competition in 2012, Alexnet won the championship with a great advantage. As a result, classification and detection methods based on deep learning have become the focus of researchers and technology companies. The subsequent emergence of VGG, GoogLeNet, and ResNet further improved the ability of convolutional neural networks to automatically extract features. Because detection based on deep learning is also a detection method for modeling targets, the rapid development of classification networks naturally also promotes the rapid development of detection models based on deep learning. In 2018, Jedom proposed the YOLOv3 model again. This model absorbed the multi-scale fusion method of ResNet's residual network structure and feature maps, enabl...
no. 2 example
[0087] In order to solve the problem that the target detection model provided in the related art cannot balance the detection accuracy and detection speed, this embodiment shows a vehicle target detection device. For details, please refer to Figure 6 , the vehicle target detection device of the present embodiment includes:
[0088] The configuration module 601 is used to cluster the size of the labeled frame of the training data set, and configure the theoretical receptive field according to the clustering result;
[0089] The first building block 602 is configured to construct a backbone network based on a theoretical receptive field; wherein, the receptive field of the output feature layer of the backbone network matches the vehicle size in a road scene;
[0090] The second building block 603 is used to configure different loss functions hierarchically for all output feature layers of the backbone network to build a vehicle detection model;
[0091] The training module 604...
no. 3 example
[0109] This embodiment provides an electronic device, see Figure 7 As shown, it includes a processor 701, a memory 702 and a communication bus 703, wherein: the communication bus 703 is used to realize connection and communication between the processor 701 and the memory 702; the processor 701 is used to execute one or more programs stored in the memory 702 A computer program to implement at least one step in the vehicle target detection method in the first embodiment above.
[0110] The present embodiment also provides a computer-readable storage medium, which includes information implemented in any method or technology for storing information, such as computer-readable instructions, data structures, computer program modules, or other data. volatile or nonvolatile, removable or non-removable media. Computer-readable storage media include but are not limited to RAM (Random Access Memory, random access memory), ROM (Read-Only Memory, read-only memory), EEPROM (Electrically Er...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More