CN113994885A - Method for seed production, cost saving, impurity removal, castration and purity preservation of hybrid corn - Google Patents
Method for seed production, cost saving, impurity removal, castration and purity preservation of hybrid corn Download PDFInfo
- Publication number
- CN113994885A CN113994885A CN202111360324.2A CN202111360324A CN113994885A CN 113994885 A CN113994885 A CN 113994885A CN 202111360324 A CN202111360324 A CN 202111360324A CN 113994885 A CN113994885 A CN 113994885A
- Authority
- CN
- China
- Prior art keywords
- corn
- hybrid
- plant
- tassel
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- A—HUMAN NECESSITIES
- A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
- A01H—NEW PLANTS OR NON-TRANSGENIC PROCESSES FOR OBTAINING THEM; PLANT REPRODUCTION BY TISSUE CULTURE TECHNIQUES
- A01H1/00—Processes for modifying genotypes ; Plants characterised by associated natural traits
- A01H1/02—Methods or apparatus for hybridisation; Artificial pollination ; Fertility
-
- A—HUMAN NECESSITIES
- A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
- A01H—NEW PLANTS OR NON-TRANSGENIC PROCESSES FOR OBTAINING THEM; PLANT REPRODUCTION BY TISSUE CULTURE TECHNIQUES
- A01H1/00—Processes for modifying genotypes ; Plants characterised by associated natural traits
- A01H1/04—Processes of selection involving genotypic or phenotypic markers; Methods of using phenotypic markers for selection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Genetics & Genomics (AREA)
- Biomedical Technology (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Biophysics (AREA)
- Artificial Intelligence (AREA)
- Botany (AREA)
- Developmental Biology & Embryology (AREA)
- Environmental Sciences (AREA)
- Breeding Of Plants And Reproduction By Means Of Culturing (AREA)
Abstract
The invention relates to a method for seed production, saving cost, removing impurities, emasculating and keeping purity of hybrid corn, which adjusts the seeding time according to the planting growth period characteristics of a male parent and a female parent and the flowering time of the male parent, so that the silking period of the female parent and the pollen dispersing period of the male parent are met; respectively carrying out hybrid identification by utilizing a corn hybrid identification model after the parent seedling setting, the node pulling and the castration of the seed production corn, and removing the hybrid; in the period of female parent tassel emergence, the maize tassel recognition model is used for recognizing the tassel, and the tassel which is extracted is pulled out; repeatedly removing the male ear plants which are not removed from the female parent, and ensuring that the male ears of the female parent are removed 100% before pollen is scattered after the female parent is castrated; in the process of producing the seeds of the corns, the labor intensity of manual impurity removal and female parent emasculation is reduced, the influence of human factors is avoided, and the purity of the hybrid corn seeds is improved.
Description
Technical Field
The invention belongs to the technical field of agriculture, and particularly relates to the field of methods for castration and purity preservation of hybrid corn seed production.
Background
Corn is one of the most important crops in the world and is also the crop where heterosis utilization is most successful. The excellent corn hybrid is an important genetic basis for high yield and high quality of corn, the purity of the hybrid is one of the measurement standards for the quality of seeds, and is an important precondition for ensuring the full exertion of the heterosis, the corn is a self-pollination crop, and emasculation is one of the most important works for hybrid seed production, so that the strict emasculation and purity preservation of female parents are required in the hybrid seed production process of the corn; the purity control of the hybrid seed production field is accompanied with the whole seed production process, the current impurity removal and female parent emasculation or artificial removal mode is taken as the main mode, and the method mainly comprises the steps of judging the characters of the leaf color, the leaf shape and the like of the seedling when the male parent and the female parent have 4 to 5 leaves in the seedling stage; before emasculation, the male ears of all maternal rows are thoroughly cleaned in time before pollen scattering or silk drawing according to agronomic characters such as plant height, stem thickness and the like of a father female parent and the female parent, so that the emasculation of a breeding field is carried out for 1 time and 2 times for rechecking, the work is continuously maintained for 15-20 days, the labor intensity is high, a large amount of labor force is needed, the experience of identifying the hybrid plants of the corn is needed, different people can influence the impurity removal effect due to different observation angles, different hybrid identification abilities and the like, the risk of exceeding the standard of seed purity is increased due to manual impurity removal and emasculation methods, the seed production cost is high, and particularly, the rejection caused by unqualified seed quality of the breeding field when the incomplete accumulated pollen scattering rate of the female parent exceeds 1 percent; the purity of the seeds is improved, the heterosis can be fully exerted, the popularization process of new varieties can be accelerated, and the yield in agricultural production is improved, so that the research and the solution of the seed purity have great and profound significance.
Disclosure of Invention
In order to solve the problems, the invention provides an intelligent method for hybrid corn seed production, which can save cost, remove impurities, castrate and keep purity by overcoming the problems of high difficulty and incomplete castration in the conventional seed production process.
The invention realizes the purpose through the following technical scheme:
a method for seed production, cost saving, emasculation and purity preservation of hybrid corn comprises the following steps;
step S1: adjusting the sowing period according to the planting growth period characteristics of the male parent and the female parent and the flowering time of the male parent, so that the silking period of the female parent meets the pollen dispersing period of the male ear of the male parent;
step S2: respectively carrying out hybrid identification by utilizing a corn hybrid identification model after the parent seedling setting, the node pulling and the castration of the seed production corn, and removing the hybrid;
step S3: in the period of female parent tassel emergence, the maize tassel recognition model is used for recognizing the tassel, and the tassel which is extracted is pulled out;
and step S4, repeating the step S3 to ensure that the tassel of the female parent is removed by 100 percent before the pollen is scattered after the female parent is castrated.
As a further optimization scheme of the present invention, the specific steps of performing hybrid identification by the corn hybrid identification model in step S2 include:
s21, shooting field corn plants by using an unmanned aerial vehicle carrying a CCD camera to generate real-time video stream data;
s22, collecting a corn plant image from the real-time video stream data, identifying the corn plant image through a hybrid identification model, and detecting whether the image has hybrid plants;
step S23, outputting the position coordinates of the hybrid plant in the image if the hybrid plant is detected, otherwise, returning to the step S22 to continue the hybrid plant detection; the position coordinates of the hybrid plants comprise the position information of the rows and the columns of the plant distribution.
As a further optimization scheme of the present invention, the specific steps of performing tassel identification through the corn tassel identification model in step S3 include:
s31, shooting field corn plants by using an unmanned aerial vehicle carrying a CCD camera to generate real-time video stream data;
s32, collecting a corn plant image from the real-time video stream data, identifying the corn plant image through a corn tassel identification model, and detecting whether the image has corn tassels;
and step S33, outputting the position coordinates of the corn tassel in the image if the corn tassel is detected, otherwise returning to the step S32 to continue the corn tassel detection.
As a further optimization scheme of the invention, after the corn plant image is collected from the real-time video stream data, the corn plant image is preprocessed by utilizing a RETINEX algorithm to eliminate the influence of the ambient light.
As a further optimization scheme of the invention, the corn hybrid identification model and the corn tassel identification model are both constructed by adopting a convolutional neural network, and the convolutional neural network adopts a YOLOV3 network.
As a further optimization scheme of the invention, the establishment of the corn hybrid identification model through the convolutional neural network specifically comprises the following steps;
a1, collecting corn plant specimen images in corn fields, including normal corn plant images and corn hybrid plant images;
a2, preprocessing the collected plant image, eliminating the influence of ambient light by using a RETINEX algorithm, and deleting the abnormal image with poor shooting effect;
a3, selecting a plant area by using a rectangular frame in the image by using a marking tool, marking the position coordinates of a normal plant and a mixed plant, and making a training set by using marked image data;
and A4, training the convolutional neural network by using the prepared training set to generate a corn hybrid identification model.
As a further optimization scheme of the invention, the corn hybrid plants in the step A1 comprise seedling hybrid plants and pre-emasculation hybrid plants after jointing, wherein the seedling hybrid plants are seedling plants with different colors, different leaf sheath colors and different leaf shapes; the hybrid plants before emasculation after jointing comprise plants with different plant types, different plant heights, different thicknesses of stems, different leaf shapes, different leaf colors, different leaf sizes, different opening angles, different tassel branches and large growth speed differences.
As a further optimized scheme of the present invention, in step S2, the hybrid removing device is used to mechanically remove the hybrid, and after the position coordinates of the hybrid in the image are obtained, the position coordinates are transmitted to the hybrid removing device, and the hybrid removing device moves to the position of the hybrid corn, so as to identify and remove the hybrid corn.
In a further preferred embodiment of the present invention, in step S3, the tassel removed is mechanically removed by a mixed plant removing device, the position coordinates of the castrated female parent are obtained and then transmitted to the mixed plant removing device, the mixed plant removing device moves to the position of the castrated female parent, the tassel of corn is identified and mechanically removed
The invention has the beneficial effects that:
1) in the process of producing the seeds of the corns, the labor intensity of manual impurity removal and female parent emasculation is reduced, the influence of human factors is avoided, and the purity of the hybrid corn seeds is improved;
2) the mechanization degree is high, the labor cost of hybrid corn seed production is obviously reduced, the improved quality of improved seeds is favorably improved, and the grain production safety is ensured.
Drawings
FIG. 1 is a flow chart of the maize hybrid identification of the present invention;
FIG. 2 is a flow chart of the corn hybrid identification model training of the present invention;
FIG. 3 is a flow chart of corn tassel identification of the present invention;
FIG. 4 is a flow chart of a corn tassel recognition model training process of the present invention;
FIG. 5 is a schematic structural diagram of a miscellaneous plant removing apparatus according to the present invention;
Detailed Description
The present application will now be described in further detail with reference to the drawings, it should be noted that the following detailed description is given for illustrative purposes only and is not to be construed as limiting the scope of the present application, as those skilled in the art will be able to make numerous insubstantial modifications and adaptations to the present application based on the above disclosure.
FIG. 1 to FIG. 5 show a cost-saving, emasculation and purity-keeping method for hybrid corn seed production; the method comprises the following steps:
step S1: adjusting the sowing period according to the planting growth period characteristics of the male parent and the female parent and the flowering time of the male parent, so that the silking period of the female parent meets the pollen dispersing period of the male ear of the male parent;
step S2: respectively carrying out hybrid identification by utilizing a corn hybrid identification model after the parent seedling setting, the node pulling and the castration of the seed production corn, and removing the hybrid;
step S3: in the period of female parent tassel emergence, the maize tassel recognition model is used for recognizing the tassel, and the tassel which is extracted is pulled out;
step S4, repeating step S3, recognizing the tassel removal condition by using the corn tassel recognition model, removing the uncleaned tassel plants in the female parent, and ensuring that 100% of the tassels of the female parent are removed within 2-4 days before pollen is scattered after the female parent is castrated;
specifically, the specific steps of performing hybrid identification by the hybrid maize identification model in step S2 include: s21, shooting field corn plants by using an unmanned aerial vehicle carrying a CCD camera to generate real-time video stream data;
s22, collecting corn plant images from real-time video stream data, compensating the collected corn plant images, eliminating illumination nonuniformity by using a RETINEX algorithm according to the influence of severe changes of environmental illumination on the images in different seasons and under different weather conditions, and keeping the consistency of brightness and color of the images so that the collected images have better stability in different illumination environments; then identifying the corn plant image through a hybrid plant identification model, and detecting whether hybrid plants exist in the image;
step S23, outputting the position coordinates of the hybrid plant in the image if the hybrid plant is detected, otherwise, returning to the step S22 to continue the hybrid plant detection; the position coordinates of the hybrid plants comprise the position information of the rows and the columns of the plant distribution;
meanwhile, the specific steps of identifying the tassel through the corn tassel identification model in the step S3 include:
s31, shooting field corn plants by using an unmanned aerial vehicle carrying a CCD camera to generate real-time video stream data;
s32, collecting a corn plant image from the real-time video stream data, identifying the corn plant image through a corn tassel identification model, and detecting whether the image has corn tassels; similarly, before the identification, the illumination nonuniformity is eliminated by utilizing a RETINEX algorithm, and the consistency of the brightness and the color of the image is kept;
step S33, if the maize tassel is detected, outputting the position coordinates of the maize tassel in the image, otherwise, returning to the step S32 to continue the maize tassel detection;
furthermore, a corn hybrid plant identification model is constructed by adopting a convolutional neural network, the convolutional neural network adopts a YOLOV3 network, the accuracy rate and the running speed are high, and the network is optimized by setting a plurality of candidate windows, so that the adaptability to plant size change can be effectively improved;
the method specifically comprises the following steps of establishing a corn hybrid identification model through a convolutional neural network;
a1, collecting corn plant specimen images in corn fields, including normal corn plant images and corn hybrid plant images; the hybrid corn plants are divided into hybrid plants in the seedling stage and hybrid plants before emasculation after jointing, the hybrid plants in the seedling stage are seedling plants with different colors, different leaf sheath colors and different leaf shapes, and the hybrid plants before emasculation after jointing comprise plants with different plant types, different plant heights, different stem thicknesses, different leaf shapes, different leaf colors, different leaf sizes, different opening angles, different tassel branches and different growth speed differences; when the images are collected, shooting the corn seedlings at a certain angle (30-90 degrees) above the plants and the ground by using an unmanned aerial vehicle before emasculation is carried out after the corn seedlings grow and the maize joints are pulled out;
the sample images to be collected for each growth period are as follows:
a2, preprocessing the collected plant image, eliminating the influence of ambient light by using a RETINEX algorithm, and deleting the abnormal image with poor shooting effect; compensating the acquired corn plant images, eliminating illumination nonuniformity by using a RETINEX algorithm according to the influence of severe changes of environmental illumination on the images under different seasons and weather conditions, keeping the consistency of the brightness and color of the images, ensuring that the acquired images have better stability under different illumination environments, and ensuring the reliability of a final training set;
a3, selecting a plant area by using a rectangular frame in the image by using a marking tool, marking the position coordinates of a normal plant and a mixed plant, and making a training set by using marked image data;
a4, training the convolutional neural network by using the manufactured training set to generate a corn hybrid plant identification model;
in addition, the corn tassel recognition model is also constructed by adopting a convolutional neural network, and in the process of constructing the corn tassel recognition model,
b1, firstly, collecting corn plant specimen images in the corn field, wherein the corn plant specimen images specifically comprise sample images of corn plants without tassel and corn tassel plants;
b2, preprocessing the collected plant images, eliminating the influence of ambient illumination by using a RETINEX algorithm, and deleting abnormal images with poor shooting effect;
b3, marking the position of the corn tassel, and making a training set by using the marked image;
b4, training the network by using a training set to generate a corn tassel recognition model;
in addition, after the corn hybrid identification model identifies the hybrid, the hybrid is removed mechanically by using a hybrid removing device, after the position coordinates of the corn hybrid in the image are obtained, the position coordinates are transmitted to the hybrid removing device, the hybrid removing device moves to the position of the corn hybrid, and the corn hybrid is identified and removed; specifically, the clearing device comprises a movable platform, a walking assembly is arranged at the bottom of the movable platform, the walking assembly can adopt walking wheels or a crawler belt, and preferably, the walking assembly adopts the crawler belt, so that the clearing device can walk in the field conveniently; the front end of the movable platform is provided with a camera, and the middle part of the movable platform is provided with at least one mechanical arm, wherein the mechanical arm adopts a multi-degree-of-freedom mechanical arm to facilitate two groups of cleaning work at the same time, so that the working efficiency is improved; a plant crushing device is arranged at the end part of the mechanical arm, and comprises a crushing cutter body for crushing the sundry plants and a gripper arranged below the crushing cutter body; when the corn plant smashing device is used, when the mixed plant removing device moves to the marked mixed plant position, the mixed plant is fixedly grabbed by the grab hand, the rod part of the corn plant extends to the smashing cutter body, and then the mixed plant is gradually smashed by the smashing cutter body;
in the stamina stage of the female parent, after the male ear of the female parent is identified by the corn tassel identification model, the tassel which is extracted by the female parent is pulled out by adopting a mixed plant removing device, after the position coordinates of the stamina-extracted corn plant in the image are obtained, the position coordinates are transmitted to the mixed plant removing device, the mixed plant removing device moves to the position of the stamina-extracted corn plant, the extracted corn tassel is identified, and the tassel is pulled out; when the corn tassel removing device is used specifically, when the mixed plant removing device moves to the marked position of the tassel-drawing corn plants, the tassel is fixedly held by the hand grip, the tassel extends to the crushing cutter body, and then the tassel is pulled out and crushed by the crushing cutter body;
example one
The hybrid corn seeds used in the embodiment are hybrid corn varieties 'full corn 1233' and male parent '512' and female parent '533'.
Planting of materials
Respectively performing direct seeding on a male parent "512" and a female parent "533" of the 'full jade 1233' in a corn seed production base in 2021 year and 4 months, and planting the seeds in 2 fields (a field A and a field B, the areas of which are 2000 square meters): wherein the male parent '512' is sown in 17 days 4 months; the female parent "533" is sown in 4 months and 15 days; the male parent and the female parent are planted in the same field (1 row of the male parent: 7 rows of the female parent), and the row spacing is 50cm multiplied by 19 cm. The field block A is a control field block, and the seed production purity is improved by adopting an artificial impurity removal and emasculation method in the hybrid seed production process; the field B is used for removing the hybrid seeds by the method and improving the seed production purity.
Secondly, rapidly identifying and removing the hybrid plants from the final singling to the flowering period
1. The male parent "533" seedlings are fixed in the field B at 18 days in 5 months, the female parent "512" seedlings are fixed at 17 days in 5 months, after the seedlings are fixed, the unmanned aerial vehicle carries a CCD camera, the height of the CCD camera is 1.5-2 meters away from the ground, the plant is vertical to the ground or forms a certain angle (30-90 degrees) with the ground to shoot the field corn seedlings, and real-time video stream data is generated; collecting a corn seedling stage image from real-time video stream data; collecting a seedling stage sample image containing normal plants and hybrid plants; the mixed plants are the mixed plants with obvious difference in leaf color, leaf sheath color and leaf shape from normal seedlings.
The method comprises the steps of preprocessing a corn seedling stage image, compensating the acquired corn plant image, eliminating the influence of ambient illumination by utilizing a RETINEX algorithm aiming at the influence of severe changes of the ambient illumination on the image under different seasons and weather conditions, mainly eliminating illumination nonuniformity by utilizing the RETINEX algorithm, keeping the consistency of the brightness and the color of the image and enabling the acquired image to have better stability under different illumination environments;
inputting the images into a hybrid plant identification model (generating the hybrid plant identification model by training a convolutional neural network, acquiring corn plant specimen images in corn fields, including normal plant and hybrid plant (seedlings with different colors, leaf sheath colors and leaf shapes), preprocessing the acquired plant images, eliminating the influence of ambient light by using a RETINEX algorithm, deleting abnormal images with poor shooting effect, marking the positions of the normal plant and the hybrid plant, making a training set by using the marked images, defining the convolutional neural network, training the network by using the training set, generating the hybrid plant identification model after the training is finished), detecting whether the hybrid plant exists in the images, outputting the position coordinates of the hybrid plant in the images when the hybrid plant is detected, finally displaying the position of the hybrid plant, and performing the correlation conformity with a corn production technology expert, determining as a hybrid plant; the self-propelled intelligent mixed plant recognition and removal machine automatically moves to a mixed plant position according to the row and column positions of the mixed plants, and corn single plant automatic crushing equipment on the side wall of the self-propelled intelligent mixed plant recognition and removal machine is used for crushing and removing the mixed plants.
2. In the field B, carrying a CCD camera by an unmanned aerial vehicle twice at a height of 3-4 m from the ground and at a position vertical to the ground or at a certain angle (30-90 degrees) with the ground in a large bell mouth period of 6 months and 25 days after jointing and in a period of 7 months and 5 days before castration to shoot field corn plants so as to generate real-time video streaming data; collecting a corn plant image from real-time video stream data; the collected corn sample image comprises normal plants and hybrid plants. The hybrid plant is a hybrid plant with obvious differences from a normal plant in plant type, plant height, stem thickness, leaf shape, leaf color, leaf size, opening angle, tassel branch number and the like.
The method comprises the steps of preprocessing a corn plant image, compensating the acquired corn plant image, eliminating the influence of ambient illumination by utilizing a RETINEX algorithm aiming at the influence of severe changes of the ambient illumination on the image under different seasons and weather conditions, mainly eliminating illumination nonuniformity by utilizing the RETINEX algorithm, keeping the consistency of the brightness and the color of the image and enabling the acquired image to have better stability under different illumination environments;
inputting the images into a hybrid plant identification model (generating the hybrid plant identification model by training a convolutional neural network, collecting corn plant specimen images in a corn field, wherein the corn plant specimen images comprise normal plants and hybrid plant (the hybrid plants with obvious differences from the normal plants in plant types, plant heights, stem thicknesses, leaf shapes, leaf colors, leaf sizes, opening angles, tassel branch numbers and the like), preprocessing the collected plant images, eliminating the influence of ambient light by using a RETINEX algorithm, deleting abnormal images with poor shooting effects, marking the positions of the normal plants and the hybrid plants, making a training set by using the marked images, defining the convolutional neural network used, training the network by using the training set, generating the hybrid plant identification model after the training is finished), detecting whether the hybrid plants exist in the images, outputting position coordinates of the hybrid plants in the images when the hybrid plants are detected, and finally displaying the positions of the hybrid plants, and the correlation of the hybrid plants is consistent with that of the technical experts in corn production, and the hybrid plants are determined; the mixed plant removing device automatically moves to the position of the mixed plant according to the row and column positions of the mixed plant, and crushing and removing the mixed plant by using crushing equipment on the side wall of the mixed plant removing machine;
3. in the tasseling period from 10 days at 7 months to 17 days at 7 months of the female parent '533', continuously shooting field corn plants in the morning and afternoon every day by using an unmanned aerial vehicle to carry a CCD camera with the height of 3-4 meters away from the ground and vertical to the ground or at a certain angle (30-90 degrees), generating real-time video stream data, collecting a corn image from the real-time video stream data, preprocessing the corn image, eliminating the influence of ambient light by using a RETINEX algorithm, inputting the image into a corn tassel recognition model (generated by training a convolutional neural network, comprising collecting corn plant specimen images in the corn field, including sample images of non-tasseled corn plants and corn tassel plants, preprocessing the collected plant images, eliminating the influence of the ambient light by using the RETINEX algorithm, deleting abnormal images with poor shooting effect, and marking the positions of the corn tassels, making a training set by using the marked image, defining a used convolutional neural network, training the network by using the training set, generating a corn tassel recognition model after the training is finished), detecting whether the corn tassel exists in the image, recognizing the length of the corn tassel and the tassel removal condition when the corn tassel is detected, outputting the position coordinate of the corn tassel in the image, and removing the tassel which is extracted by 3-5 cm by using a miscellaneous plant removing device, or crushing and removing the tassel which is detected not to be removed completely.
The field A adopts a manual impurity removal method to remove impurity plants in each period, and is emasculated by a manual emasculation method in the female parent androgenesis period.
The male parent "512" is cut off after pollination.
Thirdly, purity identification
Respectively harvesting hybrid seeds 'full jade 1233' corresponding to the field A and the field B in the corn maturity period, randomly extracting seeds of the 'full jade 1233' to germinate, randomly taking 384 leaves after one week of germination to extract DNA, and carrying out indoor SSR molecular marker identification on purity to obtain 8 hybrid strains (6 hybrid powder strains and 2 female parents) in the 384 strains of the field A, wherein the seed production purity is 97.9%; among 384 strains in the field B, 2 mixed strains (Chuanfen strain) were obtained, and the seed purity was 99.5%.
The above examples show that compared with the manual impurity removal method, the method of the invention can improve the seed production purity, realize the full-process mechanical intelligent operation, reduce the labor cost and further reduce the seed production cost.
In the process of producing the seeds of the corns, the labor intensity of manual impurity removal and female parent emasculation is reduced, the influence of human factors is avoided, and the purity of the hybrid corn seeds is improved; and the degree of mechanization is high, the labor cost of hybrid corn seed production is obviously reduced, the improved quality of improved seeds is favorably improved, and the safety of grain production is ensured.
The above-mentioned embodiments only express several embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention.
Claims (9)
1. A method for seed production, cost saving, emasculation and purity preservation of hybrid corn is characterized in that: comprises the following steps;
step S1: adjusting the sowing period according to the planting growth period characteristics of the male parent and the female parent and the flowering time of the male parent, so that the silking period of the female parent meets the pollen dispersing period of the male ear of the male parent;
step S2: respectively carrying out hybrid identification by utilizing a corn hybrid identification model after the parent seedling setting, the node pulling and the castration of the seed production corn, and removing the hybrid;
step S3: in the period of female parent tassel emergence, the maize tassel recognition model is used for recognizing the tassel, and the tassel which is extracted is pulled out;
and step S4, repeating the step S3 to ensure that the tassel of the female parent is removed by 100 percent before the pollen is scattered after the female parent is castrated.
2. The method for seed production, cost saving, emasculation and purity preservation of hybrid corn according to claim 1, which is characterized in that: the specific steps of the corn hybrid identification model in the step S2 for hybrid identification include:
s21, shooting field corn plants by using an unmanned aerial vehicle carrying a CCD camera to generate real-time video stream data;
s22, collecting a corn plant image from the real-time video stream data, identifying the corn plant image through a hybrid identification model, and detecting whether the image has hybrid plants;
step S23, outputting the position coordinates of the hybrid plant in the image if the hybrid plant is detected, otherwise, returning to the step S22 to continue the hybrid plant detection; the position coordinates of the hybrid plants comprise the position information of the rows and the columns of the plant distribution.
3. The method for seed production, cost saving, emasculation and purity preservation of hybrid corn according to claim 2, characterized in that: the specific steps of identifying the tassel through the corn tassel identification model in the step S3 include:
s31, shooting field corn plants by using an unmanned aerial vehicle carrying a CCD camera to generate real-time video stream data;
s32, collecting a corn plant image from the real-time video stream data, identifying the corn plant image through a corn tassel identification model, and detecting whether the image has corn tassels;
and step S33, outputting the position coordinates of the corn tassel in the image if the corn tassel is detected, otherwise returning to the step S32 to continue the corn tassel detection.
4. The cost-saving, emasculation and purity-maintaining method for hybrid corn seed production according to claim 2 or 3, characterized in that: after the corn plant images are collected from real-time video stream data, the corn plant images are preprocessed by utilizing a RETINEX algorithm to eliminate the influence of environmental illumination.
5. The cost-saving, emasculation and purity-maintaining method for hybrid corn seed production according to claim 4, characterized in that: the corn hybrid identification model and the corn tassel identification model are both constructed by adopting a convolutional neural network, and the convolutional neural network adopts a YOLOV3 network.
6. The method for seed production, cost saving, emasculation and purity preservation of hybrid corn according to claim 5, characterized in that: the method specifically comprises the following steps of establishing a corn hybrid identification model through a convolutional neural network;
a1, collecting corn plant specimen images in corn fields, including normal corn plant images and corn hybrid plant images;
a2, preprocessing the collected plant image, eliminating the influence of ambient light by using a RETINEX algorithm, and deleting the abnormal image with poor shooting effect;
a3, selecting a plant area by using a rectangular frame in the image by using a marking tool, marking the position coordinates of a normal plant and a mixed plant, and making a training set by using marked image data;
and A4, training the convolutional neural network by using the prepared training set to generate a corn hybrid identification model.
7. The method for seed production, cost saving, emasculation and purity preservation of hybrid corn according to claim 6, characterized in that: the hybrid corn plants in A1 comprise hybrid plants in the seedling stage and hybrid plants before emasculation after jointing, wherein the hybrid plants in the seedling stage are seedling plants with different colors, different leaf sheath colors and different leaf shapes; the hybrid plants before emasculation after jointing comprise plants with different plant types, different plant heights, different thicknesses of stems, different leaf shapes, different leaf colors, different leaf sizes, different opening angles, different tassel branches and large growth speed differences.
8. The method for seed production, cost saving, emasculation and purity preservation of hybrid corn according to claim 7, which is characterized in that: in the step S2, the hybrid removing device is used to mechanically remove the hybrid, and after the position coordinates of the hybrid in the image are obtained, the position coordinates are transmitted to the hybrid removing device, and the hybrid removing device moves to the position of the corn hybrid to identify and remove the corn hybrid.
9. The method for seed production, cost saving, emasculation and purity preservation of hybrid corn according to claim 7, which is characterized in that: in the step S3, the tassel removed is mechanically removed by using a mixed plant removing device, after the position coordinates of the castrated female parent are obtained, the position coordinates are transmitted to the mixed plant removing device, the mixed plant removing device moves to the position of the castrated female parent, and the tassel of corn is recognized and mechanically removed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111360324.2A CN113994885A (en) | 2021-11-17 | 2021-11-17 | Method for seed production, cost saving, impurity removal, castration and purity preservation of hybrid corn |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111360324.2A CN113994885A (en) | 2021-11-17 | 2021-11-17 | Method for seed production, cost saving, impurity removal, castration and purity preservation of hybrid corn |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113994885A true CN113994885A (en) | 2022-02-01 |
Family
ID=79929478
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111360324.2A Pending CN113994885A (en) | 2021-11-17 | 2021-11-17 | Method for seed production, cost saving, impurity removal, castration and purity preservation of hybrid corn |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113994885A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114766347A (en) * | 2022-04-26 | 2022-07-22 | 新疆荣耀九天科技有限公司 | Method for acquiring target three-dimensional position of corn tasseling operation |
| CN116098056A (en) * | 2023-04-13 | 2023-05-12 | 北京市农林科学院智能装备技术研究中心 | Corn emasculation inspection robot and emasculation method using same |
| CN116098055A (en) * | 2023-04-13 | 2023-05-12 | 北京市农林科学院智能装备技术研究中心 | Corn emasculation and weak removal robot and method for emasculation and weak removal by using same |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101313664A (en) * | 2008-07-07 | 2008-12-03 | 丹东登海良玉种业有限公司 | Process for preparing corn hybrid Liangyu 88 |
| CN106359074A (en) * | 2016-08-24 | 2017-02-01 | 张掖市农业科学研究院 | One-male-in-two-female corn seed production method |
| CN111837593A (en) * | 2020-07-24 | 2020-10-30 | 安徽农业大学 | Novel peucedanum praeruptorum weeding machine based on machine vision and convolutional neural network algorithm |
-
2021
- 2021-11-17 CN CN202111360324.2A patent/CN113994885A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101313664A (en) * | 2008-07-07 | 2008-12-03 | 丹东登海良玉种业有限公司 | Process for preparing corn hybrid Liangyu 88 |
| CN106359074A (en) * | 2016-08-24 | 2017-02-01 | 张掖市农业科学研究院 | One-male-in-two-female corn seed production method |
| CN111837593A (en) * | 2020-07-24 | 2020-10-30 | 安徽农业大学 | Novel peucedanum praeruptorum weeding machine based on machine vision and convolutional neural network algorithm |
Non-Patent Citations (2)
| Title |
|---|
| 吴多瑛: "提高玉米杂交制种种子质量的技术措施", 《农业工程技术》 * |
| 朱红艳: "伊犁玉米机械化去雄杂交制种技术", 《上海蔬菜》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114766347A (en) * | 2022-04-26 | 2022-07-22 | 新疆荣耀九天科技有限公司 | Method for acquiring target three-dimensional position of corn tasseling operation |
| CN116098056A (en) * | 2023-04-13 | 2023-05-12 | 北京市农林科学院智能装备技术研究中心 | Corn emasculation inspection robot and emasculation method using same |
| CN116098055A (en) * | 2023-04-13 | 2023-05-12 | 北京市农林科学院智能装备技术研究中心 | Corn emasculation and weak removal robot and method for emasculation and weak removal by using same |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113994885A (en) | Method for seed production, cost saving, impurity removal, castration and purity preservation of hybrid corn | |
| CN1018516B (en) | Improved route to hybrid rice production | |
| CN108235904B (en) | Seed property maintaining and stock seed production method of early-maturing cotton variety | |
| CN110073968A (en) | A kind of fast breeding method of high oily salt tolerant peanut | |
| CN104472157A (en) | Cultivation method for new okra varieties | |
| CN110119901A (en) | A kind of planting asparagus comprehensive management platform and its management method | |
| CN103155772B (en) | Propagation method of new sunflower number 19 restorer 654 R for oil | |
| CN112005876B (en) | Method for breeding disease and insect resistant high quality cotton variety by using cotton nuclear sterile hybrid | |
| CN116391615A (en) | Novel breeding method of wheat variety | |
| CN106718850A (en) | A kind of breeding method of muskmelon cenospecies | |
| CN110915450A (en) | Method for breeding high-yield shinyleaf yellowhorn strain | |
| CN101861828B (en) | Pineapple hybridization method | |
| CN114140702A (en) | Method for removing sundries and improving seed production purity in hybrid rice seed production field | |
| CN112616648A (en) | Method for optimizing sweet potato group hybridization scheme and application thereof | |
| CN111264379A (en) | Banana cultivation method convenient for artificial pollination | |
| Genaidy et al. | Increase the economic value of the jojoba (Simmondsia chinensis) yield using evaluation of distinctive clones grown under the Egyptian environmental conditions. | |
| CN105638451A (en) | Breeding method of low-temperature-resisting and weak-light-resisting female lines of Haiyang white cucumbers and application of breeding method | |
| Pratomo et al. | Digital identification approach to characterize Hevea brasiliensis leaves | |
| CN105075852A (en) | Method for hybridizing and breeding water dropwort | |
| CN117158307A (en) | Multi-spike new line cross breeding skill | |
| CN116349600B (en) | Oat glume-cutting-free hybridization method and application | |
| CN115119746B (en) | Screening and identifying method of lodging-resistant rice sterile line and application of screening and identifying method in breeding | |
| CN111990248B (en) | Breeding method for cultivating novel ornamental dwarf sugarcane germplasm | |
| Deng et al. | Pomegranate: an emerging fruit crop in southeast United States? | |
| Németh | Parasitic production of ergot alkaloids |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |