Target detection method based on first-order gradient neural network
A neural network and target detection technology, which is applied in neural learning methods, biological neural network models, neural architectures, etc., can solve problems such as personal safety hazards, unbalanced speed and accuracy, and target detection delays, achieving low operating power consumption, The effect of balancing speed and accuracy and high recognition accuracy
Pending Publication Date: 2020-10-27
REDNOVA INNOVATIONS INC
View PDF0 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0003] Existing automatic driving target detection algorithms, such as SSD, YOLO algorithm, and faster-RCNN cannot balance speed and accuracy, and each type of detection method has more or less problems
In the actual driving process, once the target detection is delayed or inaccurate, it will cause great harm to personal safety
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 1
[0052] Such as figure 1 As shown, the present invention is a kind of target detection method based on first-order gradient neural network, and this method specifically comprises the following steps:
[0053] Step 1: Capture live images during vehicle driving;
[0054] Step 2: Convert the step 1 image to grayscale and smooth to reduce high frequency noise;
[0055] Step 3: Extract the gradient size of the step 2 image;
[0056] Step 4: On the basis of step 3, through the first-order gradient neural network, the gradient size of the first-order gradient neural network is extracted;
[0057] Step 5: Perform feature fusion of the gradient size of step 3 and the gradient size of the first-order gradient neural network in step 4;
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 a target detection method based on a first-order gradient neural network, and relates to the field of intelligent identification. The method specifically comprises the following steps: capturing an on-site image in a vehicle driving process; converting the image into gray scale and performing smoothing to reduce high-frequency noise; carrying out gradient size extraction onthe image; extracting the gradient of the first-order gradient neural network through the first-order gradient neural network; performing feature fusion on the gradient size and the gradient size ofthe first-order gradient neural network; and enabling the image after feature fusion to sequentially enter a convolutional layer Conv1, eight first-order gradient neural network convolutional layers Fire-modules and a convolutional layer Conv10, and then enter a softmax classifier, so that a target detection result can be output. The method is applied to the field of automatic driving, achieves lightweight target detection, is low in power consumption and high in recognition precision, balances the speed and precision, and enables an automatic driving technology which is a technology closely related to life to have a better safety guarantee.
Description
technical field [0001] The invention relates to the field of intelligent recognition, in particular to a target detection method based on a first-order gradient neural network. Background technique [0002] With the development of intelligent recognition, automatic driving technology has received more and more attention. This technology, which is closely related to life, needs to achieve fast and accurate target detection. [0003] Existing automatic driving target detection algorithms, such as SSD, YOLO algorithm, and faster-RCNN cannot balance speed and accuracy, and each type of detection method has more or less problems. In the actual driving process, once the target detection is delayed or inaccurate, it will cause great harm to personal safety. Therefore, it is necessary to construct a target detection method that balances speed and accuracy, so that autonomous driving technology, a technology closely related to life, is more secure. Contents of the invention [00...
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 IPC IPC(8): G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06V2201/07G06N3/047G06N3/045G06F18/253
Inventor 王堃王铭宇吴晨
Owner REDNOVA INNOVATIONS INC



