Guangdong Polytechnic Normal University

Yiqing Fu

Yu Lin Tang

Jiahao Li

Huasheng Huang

Jiepeng Yang

Qiwei Guo

Shaoming Luo

Chaojun Hou

Jiajun Zhuang

Aimin Miao

Xincai Zhuang

Jinfei Zhao

Weizhao Chen

Xuan Chu

Xiaodi Zhang

Zhiping Tan

Xing Zhu

The present invention relates to the technical field of field crop cultivation, more particularly to a training method, an evaluation method, an electronic device and a storage medium. According to the present invention, a multispectral three-dimensional point cloud map is obtained through depth information and multispectral information, and the multispectral three-dimensional point cloud map is analyzed by utilizing an FVNet three-dimensional target detection algorithm, thereby acquiring crop feature information. Thus, more comprehensive crop information can be obtained, and a crop state evaluation model constructed based on an artificial neural network can be further trained with the crop feature information.

Training method, evaluation method, electronic device and storage medium

A fruit picking method based on a three-dimensional parameter prediction model for a fruit comprises: performing first-time image acquisition processing on a to-be-picked fruit to obtain a first image; determining a first range; controlling a manipulator to perform first-time moving processing; performing intermittent gas injection treatment to lead to forced vibration of the to-be-picked fruit; performing second-time image acquisition processing many times to obtain a plurality of second images; screening out, by taking the first image as an reference object, two appointed second images deviating from an equilibrium position to the maximum extent; jointly inputting the images into a preset three-dimensional parameter prediction model for the fruit so as to obtain predicted three-dimensional parameters; controlling the manipulator to perform second-time moving processing; and performing cutting processing on a fruit stem position to make the to-be-picked fruit fall onto the manipulator.

Fruit picking method based on three-dimensional parameter prediction model for fruit

Disclosed is a fruit picking method based on a visual servo control robot, comprising: placing a throwing apparatus and a fixed photosensitive device at a first position to obtain a fixed photosensitive image; generating a first throwing path, a second throwing path, and a third throwing path; arranging recovery apparatuses; performing simultaneous rotational throwing processing to throw a first wireless photosensitive device, a second wireless photosensitive device, and a third wireless photosensitive device; receiving a first photosensitive image sequence and a second photosensitive image sequence of the first wireless photosensitive device and the second wireless photosensitive device; receiving a third photosensitive image sequence of the third wireless photosensitive device; generating a spatial position of a fruit on a to-be-picked fruit tree; and performing fruit picking processing using the visual servo control robot according to the spatial position of the fruit.

Fruit picking method based on visual servo control robot

A citrus identification method comprises: performing first-time image acquisition processing on a target citrus tree to obtain a first image; inputting the first image into a first citrus fruit identification model to be processed to obtain a first identification result sequence; performing area interception processing on the first image to obtain a citrus fruit area; obtaining roundness integrity numerical values; selecting an appointed roundness integrity numerical value, and acquiring a defect position of an appointed citrus fruit in the first image; determining a first spatial range and a second spatial range; performing both first spray injection treatment and second spray injection treatment; performing second-time image acquisition processing to obtain a second image; inputting the second image into a second citrus fruit identification model to be processed to obtain a second identification result; and generating a citrus fruit identification result.

Citrus identification method based on roundness integrity correction

Disclosed is a method for detecting an infection stage of anthracnose pathogen with pre-analysis capacity, comprising: obtaining a plurality of sample sensing data sequences; obtaining a sample citrus leaf image; obtaining a first prediction result; if the first prediction result is that the sample citrus crop is not infected by anthracnose, obtaining sample Raman spectral data and sample hyperspectral data; obtaining a first judgment result, and obtaining a second judgment result; performing labeling to obtain second training data; training a neural network model to obtain a second anthracnose prediction model; obtaining a plurality of to-be-analyzed sensing data sequences; obtaining a to-be-analyzed citrus leaf image; obtaining a second prediction result; if the second prediction result is that the to-be-analyzed citrus crop is not infected by anthracnose, obtaining a third prediction result.

Method for detecting infection stage of anthracnose pathogenic with pre-analysis capacity

Disclosed is an obstacle avoiding method and apparatus for an unmanned aerial vehicle based on a multi-signal acquisition and route planning model. The method comprises: conducting signal acquisition processing on a first environmental area to obtain an initial millimeter-wave radar signal, an initial laser radar signal, an initial image signal and an initial ultrasonic signal; generating an initial three-dimensional environmental model according to a preset dynamic environment real-time modeling method; acquiring a motion parameter and a body shape parameter of the unmanned aerial vehicle and inputting the parameters into an initial route planning model corresponding to the initial three-dimensional environmental model based on a genetic algorithm to process so as to obtain an output of the initial route planning model; judging whether the output is capable of avoiding an obstacle; if yes, generating an obstacle avoiding flight instruction to require the unmanned aerial vehicle to fly through the first environmental area.

Obstacle avoiding method and apparatus for unmanned aerial vehicle based on multi-signal acquisition and route planning model

The present application discloses a multilevel cooperation-based unmanned aerial vehicle spraying operation method. The method includes: acquiring a first sensing signal set; inputting the first sensing signal set into a prescription map generation model to acquire a first prescription map; generating the type of liquid medicine and corresponding quantity required by a to-be-treated plant, and selecting a specified unmanned aerial vehicle; transmitting a start instruction to make the specified unmanned aerial vehicle fly to an initial position; acquiring a second sensing signal set; inputting a motion parameter and a position parameter of the unmanned aerial vehicle, the remaining liquid storage quantity of the plurality of liquid medicine storage tanks and the second sensing signal set into a spraying strategy generation module for processing to acquire a first spraying strategy; performing aerial spraying operation performing cooperative ground spraying operation to realize multilevel fine spraying operation based on the prescription map.

Multilevel cooperation-based unmanned aerial vehicle spraying operation method and device

A method for improving health of an elderly person based on a 5G signal transmission technology includes: acquiring by a real terminal a first sensing signal set; modifying the first sensing signal set to obtain a second sensing signal set; transmitting the second sensing signal set to a predetermined intermediate terminal according to a predetermined 5G signal transmission technology; correspondingly transmitting by the intermediate terminal the second sensing signal set to a predetermined simulation apparatus, acquiring by the intermediate terminal physiological feature signals to obtain a physiological feature signal set; if the physiological feature signal set does not exceed a preset standard, then transmitting the second sensing signal set to an elderly-person terminal; correspondingly transmitting by the elderly-person terminal the second sensing signal set to a second restorer cluster.

Growing community of inventors

Guangzhou, China

Yiqing Fu