CN-121998906-A - Seed metering quality detection method, device, equipment, medium and product

CN121998906ACN 121998906 ACN121998906 ACN 121998906ACN-121998906-A

Abstract

The application relates to a seed metering quality detection method, a seed metering quality detection device, seed metering quality detection equipment, a seed metering medium and a seed metering product. The seed sowing quality detection method comprises the steps of obtaining a seed image sequence, determining position information sets corresponding to a plurality of seeds respectively from the seed image sequence, wherein the seeds are different seeds, the position information sets corresponding to each seed comprise the position information of the seeds in at least three seed images, determining the landing position information and the landing time information of the seeds according to the position information sets corresponding to the seeds and the acquisition time information corresponding to the seed images from which the position information in the position information sets is derived, determining the seed sowing distance between adjacent seeds according to the landing position information and the landing time information of each seed, and determining the seed sowing quality detection result according to the seed sowing distance between the adjacent seeds. By adopting the method, the seed metering quality detection precision can be improved.

Inventors

Han Kangle
CHEN FUGUI
WANG HAO
ZHU YUEBIN
ZHANG RUIXUE
HAN HONGYU
CUI ZONGZE
WANG JIN
LIU TIE
Hou Tingchen
Lv hang
Xiang Qunjie
DENG YU

Assignees

黑龙江德沃科技开发有限公司

Dates

Publication Date: 20260508
Application Date: 20251223

Claims (10)

1. A seed metering quality detection method, the method comprising: acquiring a seed image sequence, wherein seed images in the seed image sequence are acquired in a seed sowing process; determining position information sets respectively corresponding to a plurality of seeds from the seed image sequence, wherein the seeds are different seeds, and the position information set corresponding to each seed comprises the position information of the seed in at least three seed images; For each seed, determining landing position information and landing time information of the seed according to a position information set corresponding to the seed and acquisition time information corresponding to a seed image from which the position information in the position information set is derived; And determining seed metering intervals between adjacent seeds according to the landing position information and the landing time information of each seed, and determining seed metering quality detection results according to the seed metering intervals between the adjacent seeds.
2. The method of claim 1, wherein determining a set of location information for each of a plurality of seeds from the sequence of seed images comprises: For each seed image, carrying out seed identification on the seed image to obtain a seed identification result corresponding to the seed image, wherein the seed identification result comprises position information corresponding to seeds in the seed image; Determining a plurality of seed combinations for adjacent first and second seed images in the sequence of seed images, each seed combination comprising one seed in the first seed image and one seed in the second seed image; For each seed combination, determining a pixel distance between two seeds in the seed combination according to the position information; determining two seeds in the seed combination as the same seed under the condition that the pixel distance is smaller than a distance threshold value; and obtaining a position information set corresponding to the seed according to the position information corresponding to the same seed.
3. The method of claim 2, wherein the step of performing seed recognition on the seed image to obtain a seed recognition result corresponding to the seed image comprises: Converting the seed image into a binary image, and preprocessing the binary image; identifying connected areas in the preprocessed binary image, and calculating the area of each connected area; Determining the connected region with the area larger than an area threshold as a seed, and acquiring the position information of the connected region as the position information of the seed; and obtaining a seed identification result corresponding to the seed image according to the position information of the seed.
4. A method according to any one of claims 1-3, wherein determining landing position information and landing time information of the seed according to the set of position information corresponding to the seed and acquisition time information corresponding to a seed image from which the position information in the set of position information is derived, comprises: Fitting the motion track of the seed according to the position information set corresponding to the seed and the acquisition time information corresponding to the seed image from which the position information in the position information set is derived, so as to obtain initial position information, first axial speed, second axial initial speed and second axial acceleration of the seed; And acquiring a ground image position, and calculating landing position information and landing time information of the seeds according to the ground image position, the initial position information, the first axial speed, the second axial initial speed and the second axial acceleration.
5. A method according to any one of claims 1to 3, wherein the position information includes a lateral position, and wherein the determining a seed arrangement pitch between adjacent seeds based on the landing position information and the landing time information of each seed includes: Sequencing the transverse positions of each seed according to the landing time information of each seed, and taking the difference value of the transverse positions between adjacent seeds as the initial interval between the adjacent seeds; determining a time difference value between adjacent seeds according to the landing time information of each seed, and determining a compensation interval between the adjacent seeds according to the vehicle speed and the time difference value; and determining the seed sowing distance between adjacent seeds according to the initial distance and the compensation distance.
6. A method according to any one of claims 1 to 3, wherein said determining a seeding quality detection result from a seeding spacing between adjacent seeds comprises: counting the total number of the spacing related to the seed sowing spacing; screening seed sowing intervals between adjacent seeds according to the interval range, and counting the target interval number of the seed sowing intervals meeting the interval range; And calculating the seed metering qualification rate according to the total number of the intervals and the target interval number, and determining a seed metering quality detection result according to the seed metering qualification rate.
7. A seed metering quality detection apparatus, the apparatus comprising: the seed sowing device comprises an acquisition module, a seed sowing module and a seed sowing module, wherein the acquisition module is used for acquiring a seed image sequence, and seed images in the seed image sequence are acquired in the seed sowing process; the matching module is used for determining position information sets respectively corresponding to a plurality of seeds from the seed image sequence, the seeds are different seeds, and the position information set corresponding to each seed comprises the position information of the seed in at least three seed images; The calculating module is used for determining the landing position information and the landing time information of each seed according to the position information set corresponding to the seed and the acquisition time information corresponding to the seed image from which the position information in the position information set is derived; The detection module is used for determining the seed metering distance between adjacent seeds according to the landing position information and the landing time information of each seed, and determining a seed metering quality detection result according to the seed metering distance between the adjacent seeds.
8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 6 when the computer program is executed.
9. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 6.
10. A computer program product comprising a computer program, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 6.

Description

Seed metering quality detection method, device, equipment, medium and product Technical Field The application relates to the technical field of seed metering device detection, in particular to a seed metering quality detection method, a seed metering quality detection device, seed metering quality detection equipment, a seed metering medium and a seed metering product. Background With the development of intelligent agricultural machinery technology, computer vision detection technology has been gradually applied to the field of agricultural equipment performance detection due to the characteristics of non-contact measurement, strong real-time performance, comprehensive information acquisition and the like, wherein the seed metering quality detection of an air suction type seed metering device is used as a key link for guaranteeing accurate seeding, and higher requirements are put forward on the precision of the detection technology. In the prior art, a conveyor belt type detection device is mainly adopted for detecting the seed sowing quality of the air suction type seed sowing device, and the core of the device is that seeds discharged by the seed sowing device are guided to a conveyor belt which is coated with hydraulic oil in advance, the interval state of the seeds in sowing is fixed by using the viscosity of the hydraulic oil, then a camera scans seed images on the surface of the conveyor belt, and information such as the positions and intervals of the seeds is extracted to finish quality assessment. However, the conventional detection method cannot accurately detect the seed metering quality of the seed metering device while reducing the seed loss. Disclosure of Invention In view of the foregoing, it is desirable to provide a seed-metering quality detection method, apparatus, device, medium, and product that can improve seed-metering quality detection accuracy. In a first aspect, the present application provides a method for detecting seed metering quality, including: Acquiring a seed image sequence, wherein seed images in the seed image sequence are acquired in the seed sowing process; determining position information sets corresponding to a plurality of seeds respectively from a seed image sequence, wherein the seeds are different seeds, and the position information set corresponding to each seed comprises the position information of the seed in at least three seed images; For each seed, determining the landing position information and the landing time information of the seed according to the position information set corresponding to the seed and the acquisition time information corresponding to the seed image from which the position information in the position information set is derived; And determining seed metering intervals between adjacent seeds according to the landing position information and the landing time information of each seed, and determining seed metering quality detection results according to the seed metering intervals between the adjacent seeds. In one embodiment, a plurality of seed corresponding position information sets are determined from a seed image sequence, wherein the seed image is subjected to seed identification to obtain a seed identification result corresponding to the seed image according to each seed image, the seed identification result comprises seed corresponding position information in the seed image, a plurality of seed combinations are determined according to adjacent first seed images and second seed images in the seed image sequence, each seed combination comprises one seed in the first seed image and one seed in the second seed image, the pixel distance between two seeds in the seed combination is determined according to the position information according to each seed combination, the two seeds in the seed combination are determined to be the same seed under the condition that the pixel distance is smaller than a distance threshold value, and the seed corresponding position information set is obtained according to each position information corresponding to the same seed. In one embodiment, seed identification is performed on a seed image to obtain a seed identification result corresponding to the seed image, wherein the seed identification result comprises the steps of converting the seed image into a binary image, preprocessing the binary image, identifying connected areas in the preprocessed binary image, calculating the area of each connected area, determining the connected area with the area larger than an area threshold as a seed, acquiring position information of the connected area as position information of the seed, and acquiring the seed identification result corresponding to the seed image according to the position information of the seed. In one embodiment, the method comprises the steps of determining landing position information and landing time information of seeds according to a position information set corresponding to the seeds and acquisit