Intelligent vase system, flower recognition and presentation method and electronic apparatus

Patent metadata

Field	Value
Publication number	US-12229682-B2
Application number	US-202117629354-A
Country	US
Kind code	B2
Filing date	Apr 13, 2021
Priority date	Apr 28, 2020
Publication date	Feb 18, 2025
Grant date	Feb 18, 2025

Abstract

One or more embodiments of the present disclosure provide an intelligent vase system, a flower recognition and presentation method for an intelligent vase system, and an electronic apparatus. The intelligent vase system includes: a vase body; a flower recognition device configured to: acquire a flower image of a flower inserted into the vase body, perform an object extraction on the flower image to obtain at least one object flower image, perform a flower recognition on the object flower image to obtain a flower recognition result, and acquire presentation information of an object flower based on the flower recognition result; and a display device, which is on the vase body, in communication with the flower recognition device, and configured to display the presentation information.

First claim

What is claimed is:

1. An intelligent vase system, comprising:
a vase body;
a flower recognition device configured to:
acquire a flower image of a flower inserted into the vase body;
perform an object extraction on the flower image to obtain at least one object flower image;
perform a flower recognition on the object flower image to obtain a flower recognition result; and
acquire presentation information of an object flower based on the flower recognition result; and
a display device, which is on the vase body, in communication with the flower recognition device, and configured to display the presentation information.

2. The intelligent vase system of claim 1 , further comprising a camera, which is on the vase body, in communication with the flower recognition device, and configured to capture the flower image of the flower inserted into the vase body and transmit the flower image to the flower recognition device.

3. The intelligent vase system of claim 1 , wherein the flower recognition device is further configured to:
process the flower image to obtain a gray image;
perform a threshold segmentation and a binarization processing on the gray image to obtain a binary image;
acquire at least one connected domain in the binary image; and
calculate a central position coordinate of the connected domain, and in the flower image, mark an area of the object flower image relative to the flower image based on the central position coordinate.

4. The intelligent vase system of claim 3 , wherein the flower recognition device is further configured to: segment the gray image according to an optimal threshold.

5. The intelligent vase system of claim 4 , wherein to obtain the optimal threshold, the flower recognition device is further configured to:
segment the gray image into an object area and a background area according to an initial threshold;
calculate an object area pixel gray average value of the object area, a background area pixel gray average value of the background area, and a full image pixel gray average value of the gray image;
calculate a pixel variance between the object area and the background area according to the object area pixel gray average value, the background area pixel gray average value and the full image pixel gray average value; and
change the initial threshold and recalculate the pixel variance, and select a threshold with a maximum pixel variance as the optimal threshold.

6. The intelligent vase system of claim 1 , wherein the flower recognition device is further configured to:
perform the flower recognition on the object flower image in response to a confirmation instruction from a user; or
acquire the flower image again in response to a return instruction from a user, and extract the object flower image.

7. The intelligent vase system of claim 1 , wherein the flower recognition device is further configured to:
obtain the flower recognition result according to the object flower image and a pre-trained flower recognition model.

8. The intelligent vase system of claim 7 , wherein to obtain the pre-trained flower recognition model, the flower recognition device is further configured to:
acquire a sample image set and a label corresponding to each sample image in the sample image set; wherein the sample image set comprises a sample image with flowers displayed thereon; and
train a preset convolutional neural network by using a machine learning method based on the sample image set, the label corresponding to each sample image in the sample image set, a preset classification loss function and a preset back propagation algorithm, to obtain the flower recognition model.

9. The intelligent vase system of claim 8 , wherein the flower recognition device is further configured to:
acquire a plurality of initial sample images;
preprocess the initial sample images by using a data standardization method, to obtain sample images; and
construct the sample image set according to the sample images.

10. The intelligent vase system of claim 1 , wherein the presentation information comprises flower video presentation information and flower text presentation information; and
the flower recognition device is further configured to: transmit the flower video presentation information and the flower text presentation information to different display devices for display.

11. The intelligent vase system of claim 1 , further comprising:
a turntable on a bottom of the vase body; and
a motor, which is connected to the turntable, in communication with the flower recognition device, and configured to receive a rotation control instruction generated by the flower recognition device and drive the turntable to rotate based on the rotation control instruction, so as to realize rotation of the vase body.

12. The intelligent vase system of claim 1 , further comprising:
an infrared sensor inside a vase opening of the vase body, in communication with the flower recognition device, and configured to collect an insertion or pull-out state of the flower, generate a corresponding signal and transmit the signal to the flower recognition device, so that the flower recognition device generates a control signal according to the signal indicating the insertion or pull-out state to control playing of the presentation information.

13. A flower recognition and presentation method for an intelligent vase system, the intelligent vase system being the intelligent vase system of claim 1 , the flower recognition and presentation method comprising: acquiring, by the flower recognition device, the flower image of a flower inserted into the vase body;
performing, by the flower recognition device, the object extraction on the flower image to obtain at least one object flower image;
performing, by the flower recognition device, the flower recognition on the object flower image to obtain a flower recognition result;
acquiring, by the flower recognition device, presentation information of the object flower based on the flower recognition result and transmitting the presentation information to the display device; and
displaying the presentation information on the display device.

14. The flower recognition and presentation method of claim 13 , further comprising: collecting, by a camera, the flower image of the flower inserted into the vase body; and transmitting the flower image to the flower recognition device.

15. The flower recognition and presentation method of claim 13 , wherein the performing an object extraction on the flower image to obtain at least one object flower image comprises:
processing the flower image to obtain a gray image;
performing a threshold segmentation and a binarization processing on the gray image to obtain a binary image;
acquiring at least one connected domain in the binary image; and
calculating a central position coordinate of the connected domain, and in the flower image, marking an area of the object flower image relative to the flower image based on the central position coordinate.

16. The flower recognition and presentation method of claim 15 , wherein the performing a threshold segmentation on the gray image comprises: segmenting the gray image according to an optimal threshold.

17. The flower recognition and presentation method of claim 16 , further comprising:
segmenting the gray image into an object area and a background area according to an initial threshold;
calculating an object area pixel gray average value of the object area, a background area pixel gray average value of the background area, and a full image pixel gray average val

Assignees

• Boe Technology Group Co Ltd

Inventors

• Wang Ran

Classifications

• G06N3/0464

  Convolutional networks [CNN, ConvNet] · CPC title

• G06N3/09

  Supervised learning · CPC title

• G06V20/188

  Vegetation · CPC title

• G06V10/28

  Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns · CPC title

• G06T2207/20084

  Artificial neural networks [ANN] · CPC title