CN-121811261-B - Intelligent monitoring method and system for maturity of flower buds of broccoli

CN121811261BCN 121811261 BCN121811261 BCN 121811261BCN-121811261-B

Abstract

The application relates to the technical field of intelligent monitoring of plant maturity and discloses an intelligent monitoring method and system of flower bud maturity of broccoli, wherein the intelligent monitoring method comprises the steps of analyzing flower bud distribution density according to a pre-acquired broccoli image, and performing multi-angle image acquisition on a region with flower bud distribution density larger than a preset density threshold value to obtain a first image set; the method comprises the steps of extracting morphological parameters of flower buds, identifying flower bud outlines, calculating flower bud volumes, extracting morphological characteristics of areas in the flower bud outlines, calculating probability values of the flower buds belonging to preset maturity levels through characteristic fusion analysis by combining the flower bud volumes to obtain the maturity of the flower buds, simulating the growth process of the flower buds by combining the maturity of the flower buds and historical growth data, predicting the flower bud maturity time, and intelligently monitoring the flower bud maturity of the broccoli.

Inventors

ZHANG DANGSHENG
Zhong Yunkun
ZHANG KEYUE
ZHAO MIAOMIAO
TANG BIHONG
CHEN YUMIN
LIU XINGE
ZHANG NA
GUO TAO
HAO CHAO

Assignees

汉中职业技术学院

Dates

Publication Date: 20260512
Application Date: 20260306

Claims (8)

1. An intelligent monitoring method for the maturity of buds of broccoli is characterized by comprising the following steps: analyzing the distribution density of buds according to the pre-acquired cauliflower images, and performing multi-angle image acquisition on the area with the distribution density of the buds being greater than a preset density threshold value to obtain a first image set; performing time alignment and space alignment on images in the first image set, extracting morphological parameters of buds, identifying the outline of the buds, and calculating the volume of the buds; Extracting morphological characteristics of an area in the bud outline, wherein the morphological characteristics comprise curvature of each vertex in the bud outline, area ratio and distribution dispersion of a positive Gaussian curvature area and a negative Gaussian curvature area on the surface of the bud, and axial asymmetry of outlines on two sides of the bud; analyzing the distribution skewness and kurtosis of the pixel values of the bud profile, performing spectral reflection simulation on the accumulation process of the target pigment in the maturation process, and calculating a corresponding spectral index; extracting texture features of the area in the bud outline; Fusing morphological characteristics, spectrum indexes, texture characteristics and bud volumes to construct bud characteristic vectors; According to the bud feature vector, calculating the probability value that the bud belongs to a preset maturity level to obtain the maturity of the bud; and simulating the growth process of the flower buds by combining the maturity of the flower buds and historical growth data, and predicting the bud maturity time so as to intelligently monitor the maturity of the flower buds of the cauliflowers.
2. The intelligent monitoring method for the maturity of the flower buds of the broccoli according to claim 1, wherein the analyzing the distribution density of the flower buds according to the pre-acquired broccoli image, performing multi-angle image acquisition on the area with the distribution density of the flower buds being greater than the preset density threshold value to obtain a first image set, comprises: identifying the positions of the flowers and vegetables according to the pre-acquired images of the flowers and vegetables to obtain a flower bud distribution map; analyzing connected domains in the bud distribution map, taking each connected domain as a bud region, and constructing a bud region set; And according to the bud region set, performing multi-angle image acquisition on the region with the bud distribution density larger than the preset density threshold value to obtain a first image set.
3. The intelligent monitoring method for the maturity of flower buds of broccoli according to claim 2, wherein the step of performing multi-angle image acquisition on the area with the flower bud distribution density larger than the preset density threshold according to the flower bud area set to obtain a first image set comprises the following steps: According to the bud region set, calculating the bud distribution density of each region respectively, and screening out the regions with the bud distribution density larger than a preset density threshold value to obtain a first region set; analyzing the boundary range of each region in the first region set, and screening out the corresponding boundary midpoint; and calculating a corresponding acquisition angle according to the position relation between the midpoint of the boundary and a preset physical coordinate system, and acquiring multi-angle images according to the acquisition angle to obtain a first image set.
4. The intelligent monitoring method for the maturity of the flower buds of the broccoli according to claim 1, wherein the steps of performing time alignment and space alignment on the images in the first image set, extracting morphological parameters of the flower buds, identifying outline of the flower buds, and calculating volume of the flower buds comprise the following steps: Performing time alignment and space alignment on images in the first image set, extracting morphological parameters of flower buds, identifying flower bud outlines, and constructing flower bud distribution data; Calculating curvatures of different positions of the flower bud outline according to the flower bud distribution data to obtain outline characteristics, separating single flower bud data and calculating corresponding flower bud volumes.
5. The intelligent monitoring method for the maturity of flower buds of broccoli according to claim 4, wherein the steps of performing time alignment and space alignment on the images in the first image set, extracting morphological parameters of flower buds, identifying outline of the flower buds, and constructing flower bud distribution data include: Performing time alignment and space alignment on images in the first image set to obtain a corrected image set; Extracting key points and morphological parameters of each image from the corrected image set, and constructing a key point parameter set; calculating the matching degree of key point parameter sets between images, screening out two images with the maximum matching degree, and taking the two images as image references; and identifying the flower bud outline in the image standard, and sequentially adding the first image set to the corresponding distribution position to construct flower bud distribution data.
6. The intelligent monitoring method for the maturity of flower buds of a cauliflower according to claim 5, wherein calculating curvatures of different positions of flower bud outlines according to flower bud distribution data to obtain outline features, separating out single flower bud data and calculating corresponding flower bud volumes comprises: calculating curvatures of different positions of the flower bud outline according to the flower bud distribution data to obtain outline features, clustering according to the outline features, and separating out single flower bud data; projecting the data of the single flower buds onto a vertical plane respectively, and identifying the edges of the corresponding flower buds; and calculating the edge length and the edge curvature according to the edge of the flower bud, and calculating the corresponding flower bud volume through integration.
7. The intelligent monitoring method for the maturity of the flower buds of the broccoli according to claim 1, wherein the calculating the probability value of the flower buds belonging to the preset maturity level according to the flower bud feature vector to obtain the maturity of the flower buds comprises the following steps: According to the bud feature vector, analyzing the maturity of the buds through a preset neural network model, and calculating the probability value that the buds belong to a preset maturity level; and identifying the maturity of the flower buds according to the probability value.
8. An intelligent monitoring system for the maturity of the flower buds of a broccoli, which is characterized by being used for realizing the intelligent monitoring method for the maturity of the flower buds of the broccoli according to any one of claims 1-7, comprising the following steps: The image acquisition module is used for analyzing the distribution density of the flower buds according to the pre-acquired image of the broccoli, and performing multi-angle image acquisition on the area with the distribution density of the flower buds being greater than a preset density threshold value to obtain a first image set; The flower bud analysis module is used for carrying out time alignment and space alignment on the images in the first image set, extracting morphological parameters of flower buds, identifying the outline of the flower buds and calculating the volume of the flower buds; the fusion calculation module is used for extracting morphological characteristics of the area in the outline of the flower bud, calculating probability values of the flower bud belonging to a preset maturity level through characteristic fusion analysis by combining the volume of the flower bud, and obtaining the maturity of the flower bud; and the maturity monitoring module is used for simulating the growth process of the flower buds by combining the maturity of the flower buds and historical growth data, and predicting the maturity time of the flower buds so as to intelligently monitor the maturity of the flower buds of the flowers and vegetables of the tree.

Description

Intelligent monitoring method and system for maturity of flower buds of broccoli Technical Field The application relates to the technical field of intelligent monitoring of plant maturity, in particular to an intelligent monitoring method and system of the maturity of cauliflower buds. Background Currently, broccoli is an important special vegetable, and the maturity of flower buds directly influences commodity value, harvesting time and subsequent processing quality. The traditional judgment of the maturity of the flower buds of the broccoli mainly depends on farmers or quality inspection personnel with abundant experience to carry out field manual observation and hand touch feeling, and has the problems of strong subjectivity, different standards, low efficiency, incapability of large-scale real-time monitoring and the like. The intelligent monitoring method and system for the maturity of the flower buds of the tree flower are provided for solving at least one of the problems of difficulty in accurately analyzing the real shape of the flower buds of the tree flower due to the fact that the flower buds of the tree flower are densely grown, irregularly shaped and seriously shielded from each other by relying on image analysis of a single angle, high maturity judging error caused by difficulty in accurately analyzing the real shape of the flower buds of the tree flower by adopting a few dominant features such as color, size and the like, lack of analysis on the fine change of the surface textures of the flower buds, insensitivity to judgment in early maturity, lack of dynamic combination with historical growth data by adopting static analysis, incapability of simulating the growth trend of the flower buds and incapability of providing predictive decision support for an agricultural harvesting plan. Disclosure of Invention Aiming at the defects existing in the prior art, the application aims to provide an intelligent monitoring method and system for the maturity of the flower buds of broccoli, which can effectively solve the problems in the background art. The specific technical scheme of the application is as follows: An intelligent monitoring method for the maturity of the flower buds of a broccoli comprises the following steps: analyzing the distribution density of buds according to the pre-acquired cauliflower images, and performing multi-angle image acquisition on the area with the distribution density of the buds being greater than a preset density threshold value to obtain a first image set; performing time alignment and space alignment on images in the first image set, extracting morphological parameters of buds, identifying the outline of the buds, and calculating the volume of the buds; Extracting morphological characteristics of an area in the outline of the flower bud, and calculating a probability value of the flower bud belonging to a preset maturity level through characteristic fusion analysis by combining the volume of the flower bud to obtain the maturity of the flower bud; and simulating the growth process of the flower buds by combining the maturity of the flower buds and historical growth data, and predicting the bud maturity time so as to intelligently monitor the maturity of the flower buds of the cauliflowers. Specifically, according to the pre-acquired cauliflower image, analyzing the distribution density of the buds, and performing multi-angle image acquisition on an area with the distribution density of the buds being greater than a preset density threshold value to obtain a first image set, wherein the method comprises the following steps: identifying the positions of the flowers and vegetables according to the pre-acquired images of the flowers and vegetables to obtain a flower bud distribution map; analyzing connected domains in the bud distribution map, taking each connected domain as a bud region, and constructing a bud region set; And according to the bud region set, performing multi-angle image acquisition on the region with the bud distribution density larger than the preset density threshold value to obtain a first image set. Specifically, the multi-angle image acquisition is performed on the area with the flower bud distribution density greater than the preset density threshold according to the flower bud area set to obtain a first image set, which includes: According to the bud region set, calculating the bud distribution density of each region respectively, and screening out the regions with the bud distribution density larger than a preset density threshold value to obtain a first region set; analyzing the boundary range of each region in the first region set, and screening out the corresponding boundary midpoint; and calculating a corresponding acquisition angle according to the position relation between the midpoint of the boundary and a preset physical coordinate system, and acquiring multi-angle images according to the acquisition angle to obtain a first image set. S