CN-121985099-A - Intelligent monitoring method and system for edible fungus greenhouse cultivation

CN121985099ACN 121985099 ACN121985099 ACN 121985099ACN-121985099-A

Abstract

The invention relates to the technical field of video compression and provides an intelligent monitoring method and system for edible fungus greenhouse cultivation, comprising the steps of collecting edible fungus monitoring videos and uplink bandwidth data of different cameras, carrying out framing processing to obtain edible fungus monitoring images, and dividing a stipe area; the method comprises the steps of determining important frame indexes of edible fungus monitoring images, dividing edible fungus monitoring videos in seconds, calculating frame compression indexes of cameras in each second, compressing and transmitting all collected edible fungus monitoring videos according to the frame compression indexes of all cameras in each second, and monitoring edible fungus greenhouse cultivation. The invention can realize the reconstruction of the edible fungus monitoring video with optimal quality under the limited bandwidth.

Inventors

ZHANG DI
SHI LEI
LIU YUBING
Chai Fangyan
JI WEI
WANG HAIBO
JIANG GUOSHENG
XU MAN
SUI CHUNGUANG

Assignees

黑龙江农业经济职业学院
黑龙江省农业科学院牡丹江分院

Dates

Publication Date: 20260505
Application Date: 20260325

Claims (10)

1.An intelligent monitoring method for edible fungi greenhouse cultivation is characterized by comprising the following steps: Collecting edible fungus monitoring videos and uplink bandwidth data of different cameras, carrying out framing treatment on the edible fungus monitoring videos, obtaining edible fungus monitoring images, and dividing all the stipe areas in the edible fungus monitoring images according to gray value differences and texture characteristics of pixel points in the edible fungus monitoring images, wherein each edible fungus monitoring image corresponds to one uplink bandwidth data; Determining an important frame index of the edible fungus monitoring image according to the directions of the long sides of the minimum circumscribed rectangles of all the stipe areas in the edible fungus monitoring image and the distribution of pixel values in the bottom local range of the minimum circumscribed rectangles of the stipe areas; dividing an edible fungus monitoring video in seconds, and respectively calculating a frame compression index of each camera in each second according to differences of uplink bandwidth data and important frame indexes corresponding to all edible fungus monitoring images acquired by the same camera in the same second and image matching results of the edible fungus monitoring images acquired by different cameras at the same moment; And compressing and transmitting all collected edible fungus monitoring videos according to the frame compression index of all cameras in each second, so as to realize monitoring of edible fungus greenhouse cultivation.
2. The intelligent monitoring method for edible fungi greenhouse cultivation according to claim 1, wherein the dividing all the stipe areas in the edible fungi monitoring image according to the gray value difference and the texture characteristics of the pixels in the edible fungi monitoring image comprises the following specific steps: Dividing a front Jing Liantong domain according to pixel values of pixel points in the edible fungus monitoring image; The average value of the texture characteristic values of all pixel points in the foreground connected domain extracted by the texture extraction algorithm is marked as the texture characteristic average value of the foreground connected domain, and the average value of the absolute value of the difference value of the texture characteristic average value of the foreground connected domain and all other foreground connected domains is marked as the texture average difference of the foreground connected domain; Dividing the stipe area according to the average difference of textures of all foreground connected areas identified by the edible fungus monitoring images.
3. The intelligent monitoring method for edible fungi greenhouse cultivation according to claim 1, wherein the specific determining method for the important frame index of the edible fungi monitoring image is as follows: determining the average gradient of the edible fungus monitoring image according to the directions of the long sides of the minimum circumscribed rectangles of all the stipe areas in the edible fungus monitoring image; determining the average base entropy of the edible fungus monitoring image according to the distribution of pixel values in the bottom local range of the minimum circumscribed rectangle of the stipe area; and (3) marking the positive correlation processing result of the average gradient and the average base entropy of the edible fungus monitoring image as an important frame index of the edible fungus monitoring image.
4. The intelligent monitoring method for edible fungi greenhouse cultivation according to claim 3, wherein the specific acquisition method for the average gradient of the edible fungi monitoring image is as follows: And (3) recording the included angle between the direction of the long side of the minimum circumscribed rectangle of the petiole area and the vertical upward direction as the inclination of the petiole area, and recording the average value of the inclination of all the petiole areas in the same edible fungus monitoring image as the average inclination of the edible fungus monitoring image.
5. The intelligent monitoring method for edible fungi greenhouse cultivation according to claim 3, wherein the specific determination method of the average base entropy of the edible fungi monitoring image is as follows: The information entropy of gray values of all pixel points in a circle with the minimum circumscribing rectangle of the stipe area as a circle center and the first preset length as a radius is recorded as the base image entropy of the stipe area, and the average value of the base image entropy of all stipe areas in the same edible fungus monitoring image is recorded as the average base entropy of the edible fungus monitoring image.
6. The intelligent monitoring method for edible fungi greenhouse cultivation according to claim 1, wherein the method for calculating the frame compression index of the camera in each second is as follows: Determining the number of clusters corresponding to the same second according to the difference of uplink bandwidth data corresponding to all edible fungus monitoring images acquired by the same camera in the same second; calculating an overlapping index corresponding to the edible fungus monitoring images of the cameras according to image matching results of the edible fungus monitoring images acquired by the different cameras at the same time; the frame compression index of each camera in each second is calculated, and the calculation formula is as follows: In the formula, Represent the first The camera is at the first Frame compression index in seconds; represent the first The number of clusters corresponding to seconds; represent the first The camera is at the first The average value of the overlapping indexes corresponding to all the edible fungus monitoring images acquired in seconds; represent the first The camera is at the first The average value of the important frame indexes of all the edible fungus monitoring images acquired in seconds; Representing a preset parameter adjusting coefficient; Representing a hyperbolic tangent function.
7. The intelligent monitoring method for edible fungi greenhouse cultivation according to claim 6, wherein the specific determining step of the number of clusters corresponding to the same second is as follows: The variation coefficients of uplink bandwidth data corresponding to all edible fungus monitoring images of the same camera in the same second are recorded as uplink travelling power of the same camera in the same second; and clustering the uplink mobility of all cameras in the same second, and obtaining the number of clusters corresponding to the same second.
8. The intelligent monitoring method for edible fungi greenhouse cultivation of claim 6, wherein the overlapping index corresponding to the edible fungi monitoring image of the camera is specifically: any one camera is marked as a target camera, and all cameras except the target camera are marked as common network cameras; Matching edible fungus monitoring images acquired by the target camera and the common-network cameras at the same moment to acquire an overlapping area of the two edible fungus monitoring images, marking the average value of the number of pixels contained in the overlapping area of the edible fungus monitoring images acquired by the target camera and all the common-network cameras at the same moment as the average overlapping area acquired by the target camera at the same moment, and marking the ratio of the average overlapping area of the target camera to the number of pixels contained in the edible fungus monitoring images corresponding to the average overlapping area of the target camera as the overlapping index corresponding to the edible fungus monitoring images of the target camera.
9. The intelligent monitoring method for edible fungi greenhouse cultivation according to claim 1, wherein the method for compressing and transmitting all collected edible fungi monitoring videos according to the frame compression index of all cameras in each second comprises the following specific steps: obtaining an initial value of a quantization parameter of an HEVC algorithm by using a block residual error method; And taking the optimized quantization parameter of the camera in the corresponding second as the value of the quantization parameter of the HEVC algorithm, and compressing and transmitting the edible fungus monitoring video acquired by the camera in the corresponding second by using the HEVC algorithm.
10. An intelligent monitoring system for greenhouse cultivation of edible fungi, comprising a memory, a processor and a computer program stored in the memory and running on the processor, characterized in that the processor implements the steps of the method according to any of claims 1-9 when executing the computer program.

Description

Intelligent monitoring method and system for edible fungus greenhouse cultivation Technical Field The invention relates to the technical field of video compression, in particular to an intelligent monitoring method and system for greenhouse cultivation of edible fungi. Background In the edible fungus greenhouse cultivation process, the monitoring video is used for real-time monitoring, so that the accurate regulation and control of the edible fungus growth environment can be ensured, early warning, prevention and control of plant diseases and insect pests can be realized, the management efficiency is greatly improved, and the cost is reduced. In order to realize identification of edible fungus diseases, the monitoring video of edible fungus growth needs to be encoded and compressed, so that the efficiency and quality of communication transmission are improved. Wherein, the HEVC algorithm, namely the H.265 algorithm, can be used for realizing the encoding and compression of the monitoring video. But network bandwidth fluctuation and the duty ratio of an important area capable of displaying the growth condition of the edible fungi in the monitoring video can both influence the video reconstruction quality, so that the selection of quantization parameters of the HEVC algorithm is influenced, the quantization parameters of the HEVC algorithm are required to be determined according to the actual network condition and the content of the actual monitoring video, and the efficiency and the quality of transmitting the monitoring video for the growth of the edible fungi are improved. Disclosure of Invention The invention provides an intelligent monitoring method and system for edible fungus greenhouse cultivation, which aim to solve the problems that in the process of encoding monitoring videos for edible fungus growth, the video reconstruction quality is influenced by both actual network conditions and the content of the actual monitoring videos, so that the selection of quantization parameters of an HEVC algorithm is unreasonable, and the efficiency and quality of video transmission are unbalanced, and the adopted technical scheme is as follows: In a first aspect, an embodiment of the present invention provides an intelligent monitoring method for greenhouse cultivation of edible fungi, where the method includes the following steps: Collecting edible fungus monitoring videos and uplink bandwidth data of different cameras, carrying out framing treatment on the edible fungus monitoring videos, obtaining edible fungus monitoring images, and dividing all the stipe areas in the edible fungus monitoring images according to gray value differences and texture characteristics of pixel points in the edible fungus monitoring images, wherein each edible fungus monitoring image corresponds to one uplink bandwidth data; Determining an important frame index of the edible fungus monitoring image according to the directions of the long sides of the minimum circumscribed rectangles of all the stipe areas in the edible fungus monitoring image and the distribution of pixel values in the bottom local range of the minimum circumscribed rectangles of the stipe areas; dividing an edible fungus monitoring video in seconds, and respectively calculating a frame compression index of each camera in each second according to differences of uplink bandwidth data and important frame indexes corresponding to all edible fungus monitoring images acquired by the same camera in the same second and image matching results of the edible fungus monitoring images acquired by different cameras at the same moment; And compressing and transmitting all collected edible fungus monitoring videos according to the frame compression index of all cameras in each second, so as to realize monitoring of edible fungus greenhouse cultivation. Further, the dividing all the stipe areas in the edible fungus monitoring image according to the gray value difference and the texture characteristics of the pixel points in the edible fungus monitoring image comprises the following specific methods: Dividing a front Jing Liantong domain according to pixel values of pixel points in the edible fungus monitoring image; The average value of the texture characteristic values of all pixel points in the foreground connected domain extracted by the texture extraction algorithm is marked as the texture characteristic average value of the foreground connected domain, and the average value of the absolute value of the difference value of the texture characteristic average value of the foreground connected domain and all other foreground connected domains is marked as the texture average difference of the foreground connected domain; Dividing the stipe area according to the average difference of textures of all foreground connected areas identified by the edible fungus monitoring images. Further, the specific determination method of the important frame index of the edible fungi monitoring image