Potato seed preparation system
Through the image acquisition and cutting decision module of the potato seed potato preparation system, the problems of low efficiency of potato cutting and different sizes of potato pieces are solved, and automated cutting and uniformity are achieved, which is suitable for large-scale planting.
Patent Information
- Application Number
- CN202211437777.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-11-15
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2042-11-15
AI Technical Summary
During the process of cutting potatoes, there are low operating efficiency, high labor intensity, high production costs, and the cut potato pieces are of varying sizes, which cannot meet the needs of large-scale planting.
A potato seed potato preparation system is designed, including a seed cutting device, an image acquisition device and a cutting decision module. Through image acquisition and processing, a cutting strategy is generated to ensure that each of the cut potato pieces contains bud eyes and is of similar size.
It realizes the automation of potato cutting, improves cutting efficiency and uniformity of potato blocks, is suitable for large-scale sowing needs, and reduces production costs.
Smart Images

Figure CN115812380B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of agricultural planting, and in particular to a potato seed potato preparation system. Background Art
[0002] Potato is one of the fourth largest staple food crops in my country, and increasing the potato planting area is crucial to ensuring global food security. However, due to the lack of mature intelligent cutting technology in potato sowing, the yield and quality of potato production have been greatly affected, which in turn seriously affects the production benefits of farmers.
[0003] At present, potato cutting is still mainly done manually, which has problems such as low operating efficiency, high labor intensity, high production costs, and easy delays in farming season, and cannot meet the needs of large-scale potato cultivation. Summary of the invention
[0004] The purpose of the present application is to provide a potato seed potato preparation system, which is used to quickly complete the potato seed cutting operation, and each potato tuber after cutting contains bud eyes, and the sizes of different potato tubers are similar.
[0005] The present application provides a potato seed potato preparation system, comprising:
[0006] A seed cutting device, an image acquisition device and a cutting decision module; the seed cutting device comprises: a cutter and a tray, and a driving mechanism for controlling the movement of the cutter and the tray; the image acquisition device is used to acquire a target image of potatoes to be cut placed on the tray; the cutting decision module is used to generate a target cutting strategy according to the target image acquired by the image acquisition device; the driving mechanism is used to control the cutter to cut the potatoes to be cut on the tray according to the cutting scheme indicated by the target cutting strategy, so that each potato tuber of the cut potatoes contains bud eyes and the sizes of different potato tubers are similar.
[0007] Optionally, the tool comprises a plurality of blades fixed on a knife shaft; the angles between any two blades are the same; the tray is provided with equally spaced cutting grooves matching the blades, and the number of cutting grooves is an integer multiple of the number of blades; a circular hole is provided at the center of the tray, and the axis of the circular hole is in the same straight line as the knife shaft; a plurality of positioning marks are provided at the edge of the tray; the plurality of positioning marks are used to determine the target landing point of the center of the tool on the potato to be cut; the driving mechanism is specifically used to control the tool and the tray to move toward each other.
[0008] Optionally, an open slot is provided on the tray; the potato seed potato preparation system further includes: a tray detachment sensor; the drive mechanism is further used to control the tray to change its movement direction after the seed cutting device completes cutting the potatoes to be cut, so that the knife shaft slides out of the open slot; the drive mechanism is further used to control the tool and the tray to reset when the tray detachment sensor detects that the knife shaft slides out of the open slot.
[0009] Optionally, the image acquisition device includes multiple image acquisition units; each of the multiple image acquisition units is used to acquire images of the potatoes to be cut at different angles; the potato seed potato preparation system also includes: a position sensor; the position sensor is used to detect the position of the tray.
[0010] Optionally, the image acquisition device is specifically used to control the image acquisition device to acquire the target image and send the target image to the cutting decision module when the position sensor indicates that the tray is located at the first position; wherein the target image includes images of the potatoes to be cut at different angles acquired by each image acquisition unit.
[0011] Optionally, the cutting decision module is specifically configured to generate a plurality of cutting schemes according to the target image acquired by the image acquisition device, and select the target cutting scheme from the plurality of cutting schemes.
[0012] Optionally, the cutting decision module is specifically used to stitch the images in the target image after receiving the target image to generate a stitched image of the maximum viewing angle containing the potatoes to be cut; the cutting decision module is specifically used to extract the contours of the potatoes to be cut in the stitched image, and determine the target area in the stitched image containing the potatoes to be cut; the cutting decision module is specifically used to generate eye information of the potatoes to be cut based on the stitched image and the target area; the cutting decision module is specifically used to divide the target area into multiple first sub-areas along the center of each cutting groove according to the distribution of the multiple cutting grooves; the cutting decision module is specifically used to calculate the area of the potatoes to be cut covered by each first sub-area and the number of eyes contained in each first sub-area; wherein the stitched image is obtained after feature matching based on the pixel points corresponding to the positioning identifiers contained in each image in the target image as feature points.
[0013] Optionally, the cutting decision module is further used to generate a plurality of dividing lines corresponding to the number and position of the blades according to the initial positions of the plurality of blades of the tool; the cutting decision module is further used to rotate the plurality of dividing lines with the target knife landing point as the rotation center to generate a plurality of cutting schemes; one rotation generates a cutting scheme, and the angle of each rotation is the same; each segmentation scheme can divide the target area into a plurality of second sub-areas; a second sub-area covers at least one first sub-area; the number of first sub-areas covered by each second sub-area is the same; the cutting decision module is further used to calculate the area of the potatoes to be cut and the number of eyes in each second sub-area corresponding to each cutting scheme according to the first sub-area covered by each second sub-area corresponding to each cutting scheme; the cutting decision module is further used to calculate the area variance of each cutting scheme based on the area of the potatoes to be cut in each second sub-area corresponding to each cutting scheme; the cutting decision module is further used to determine the cutting scheme with the smallest area variance among the plurality of cutting schemes and containing eyes in each second sub-area as the target cutting scheme.
[0014] Optionally, the driving mechanism is specifically configured to cut the potatoes to be cut on the tray according to a cutting scheme indicated by the target cutting strategy when the position sensor indicates that the tray is located at the second position.
[0015] Optionally, the cutting decision is specifically used to identify the eyes of the potato to be cut from the spliced image and generate the eye information based on an improved lightweight target detection network; wherein the improved lightweight target detection network is improved based on the YOLOX-nano network; the improved lightweight target detection network deletes some feature layers in the backbone network of the YOLOX-nano network, and adds an attention mechanism module to the spatial pyramid pooling module of the YOLOX-nano network to form an improved backbone network; the improved lightweight target detection network replaces the feature fusion network in the YOLOX-nano network with a deep and shallow fusion feature module as a feature network to realize feature fusion.
[0016] The potato seed potato preparation system provided by the present application firstly, an image acquisition device acquires a target image of the potato to be cut placed on the tray. Then, a cutting decision module generates a target cutting strategy according to the target image acquired by the image acquisition device. Finally, a driving mechanism controls the cutter to cut the potato to be cut on the tray according to the cutting scheme indicated by the target cutting strategy, thereby realizing rapid preparation of potato seed potatoes, and at the same time making each potato tuber of the cut potato contain bud eyes, and the sizes of different potato tubers are similar. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.
[0018] Figure 1 It is a schematic diagram of the structure of a potato seed potato preparation system provided in the present application;
[0019] Figure 2 It is a schematic diagram of the structure of a tray of a potato seed potato preparation system provided in the present application;
[0020] Figure 3 It is a schematic diagram of the division of the area where potatoes are located in the spliced image by the potato seed potato preparation system provided in the present application. DETAILED DESCRIPTION
[0021] In order to make the purpose, technical solutions and advantages of this application clearer, the technical solutions in this application will be clearly and completely described below in conjunction with the drawings in this application. Obviously, the described embodiments are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0022] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by "first", "second", etc. are generally of one type, and the number of objects is not limited. For example, the first object can be one or more. In addition, "and / or" in the specification and claims represents at least one of the connected objects, and the character " / " generally indicates that the objects associated with each other are in an "or" relationship.
[0023] In the related art, the manual cutting method of potatoes has problems such as different sizes of cut pieces, poor uniformity, and failure to ensure that at least buds are left in the cut pieces, resulting in a large portion of potato pieces failing to meet sowing requirements, causing a waste of resources and increasing the cost of sowing.
[0024] In view of the above-mentioned technical problems existing in the related art, the embodiment of the present application provides a potato seed potato preparation system, which can automatically complete the cutting of potatoes. The obtained seed potatoes all contain buds and are of similar size, which are very suitable for potato sowing.
[0025] The potato seed potato preparation system provided in the embodiment of the present application is described in detail below through specific embodiments and application scenarios in conjunction with the accompanying drawings.
[0026] The present application provides a potato seed preparation system, which may include: a seed cutting device, an image acquisition device, and a cutting decision module. Figure 1 As shown, the seed cutting device includes: a cutter and a tray, and a driving mechanism for controlling the movement of the cutter and the tray.
[0027] Exemplarily, the image acquisition device is used to acquire a target image of the potatoes to be cut placed on the tray; the cutting decision module is used to generate a target cutting strategy based on the target image acquired by the image acquisition device; the driving mechanism is used to control the tool to cut the potatoes to be cut on the tray according to the cutting scheme indicated by the target cutting strategy, so that each potato tuber after cutting contains eyes and the sizes of different potato tubers are similar.
[0028] For example, Figure 1 As shown, the tool includes a plurality of blades fixed on a knife shaft; the angles between any two blades are the same; the tray is provided with equally spaced cutting grooves matching the blades, and the number of the cutting grooves is an integer multiple of the number of the blades; a circular hole is provided at the center of the tray, and the axis of the circular hole is in the same straight line as the knife shaft; a plurality of positioning marks are provided at the edge of the tray; the plurality of positioning marks are used to determine the target landing point of the center of the tool on the potato to be cut; the driving mechanism is specifically used to control the tool and the tray to move toward each other.
[0029] In the embodiment of the present application, the number of blades included in the tool is N. For example, the blades are evenly fixed on the blade shaft, and the blade line evenly divides the plane into planes of size 360° / N. The tool can rotate, move left and right, and move up and down under the action of the driving mechanism. The tray is used to place potatoes to be cut, and the middle of the tray is concave to facilitate the stable placement of potatoes during the cutting process. Figure 2 As shown, there are equally spaced cutting grooves 1 on the tray that match the cutting blades, and the angle between two adjacent cutting grooves is θ degrees and satisfies kθ=360° / N, where k is an integer. That is, the number of cutting grooves on the tray is an integer multiple of the number of blades on the tool.
[0030] For example, taking the number of cutting edges of the tool as 4 (ie, N=4), the number of cutting grooves can be 1, 2, 3...k. Assuming k=2, the angle between any two adjacent cutting grooves is θ=360° / (k×N)=45°.
[0031] It should be noted that the quantitative relationship between the cutting grooves and the blades can ensure that after the tool is rotated from the initial position to the angle indicated by the target cutting strategy, the cutting operation of the potatoes to be cut can be completed and the potatoes can pass through the cutting grooves on the tray smoothly without interfering with the cutting grooves.
[0032] In the embodiments of the present application, Figure 2 As shown, there is a circular hole 2 at the center of the tray, and the axis of the hole is in the same straight line as the axis of the cutter. When cutting, the cutter moves from top to bottom perpendicular to the ground, and the tray and the cutter move toward each other. The cutter passes through the tray from the center hole and cutting groove of the tray, and a plurality of raised positioning marks 3 are set on the edge of the tray. Using these positioning marks, the center of the cutter axis can be determined in the image captured by the image acquisition device. The point where the cutter falls on the potato.
[0033] For example, Figure 2 As shown, the tray is also provided with an opening slot 4; the potato seed preparation system further includes: a tray detachment sensor. The driving mechanism is also used to control the tray to change the movement direction after the seed cutting device completes cutting the potatoes to be cut, so that the knife shaft slides out of the opening slot; the driving mechanism is also used to control the knife and the tray to reset when the tray detachment sensor detects that the knife shaft slides out of the opening slot.
[0034] In an embodiment of the present application, the drive mechanism rotates the tool from an initial position to the angle indicated by the target cutting strategy according to the target cutting strategy generated by the cutting decision module, and the tool linear motion controller controls the cutter to move in a straight line downward, passing through the tray, and cutting the potatoes into N pieces; when the tray moves up to a certain position, the tray movement is changed to move diagonally upward, and the knife rod slides out of the tray opening slot 4, and the tray detachment sensor immediately gives a signal. After receiving the signal, the tool linear motion controller retracts the tool back to its original position, and then the tool rotation angle controller resets the tool rotation angle.
[0035] Exemplarily, the image acquisition device includes multiple image acquisition units; each of the multiple image acquisition units is used to acquire images of the potatoes to be cut at different angles; the potato seed potato preparation system also includes: a position sensor; the position sensor is used to detect the position of the tray.
[0036] In the embodiments of the present application, Figure 1As shown, taking the image acquisition device including two image acquisition units (ie, camera A and camera B) as an example, the two cameras are respectively used to acquire images of potatoes to be cut placed on a tray at different angles.
[0037] It should be noted that, based on the technical solution in the embodiments of the present application, those skilled in the art may also set a plurality of image acquisition units at a plurality of different positions to acquire images of the potatoes to be cut at various angles.
[0038] It is understandable that since the image of the potato to be cut captured from directly above cannot fully display the distribution of the eyes in different areas of the potato to be cut, two cameras at different positions are used in the embodiment of the present application to capture images of different areas of the potato to be cut.
[0039] Specifically, the image acquisition device is specifically used to control the image acquisition device to acquire the target image and send the target image to the cutting decision module when the position sensor indicates that the tray is located at the first position; wherein the target image includes images of the potatoes to be cut at different angles acquired by each image acquisition unit.
[0040] In the embodiments of the present application, Figure 1 As shown, a position sensor is installed at a certain position below the camera. When the tray moving from bottom to top passes the position sensor, the position sensor generates a first trigger signal, which is transmitted to the trigger control circuits of the two cameras to control the two cameras to take two images from two different angles. Afterwards, the collected images are transmitted to the cutting decision module to identify the buds on the potatoes to be cut.
[0041] Specifically, the cutting decision module is specifically used to, after receiving the target image, stitch the images in the target image to generate a stitched image of the maximum viewing angle of the potato to be cut;
[0042] Specifically, the cutting decision module is further configured to extract the contour of the potato to be cut in the spliced image, and determine a target area in the spliced image that contains the potato to be cut.
[0043] Specifically, the cutting decision module is further configured to generate eye information of the potato to be cut according to the spliced image and the target area.
[0044] Exemplarily, the above-mentioned eyelet information may include the number of eyelets, and the position information of each eyelet, etc.
[0045] Specifically, the cutting decision module is further configured to divide the target area into a plurality of first sub-areas along the center of each cutting groove according to the distribution of the plurality of cutting grooves.
[0046] In a possible implementation manner, the number of the plurality of first sub-regions is the same as the number of the cutting grooves.
[0047] It can be understood that the number multiples of the cutting grooves and the blades determine the number of cutting schemes. The larger the number multiples, the more cutting schemes that can be realized.
[0048] Specifically, the cutting decision module is further configured to calculate the area of the potatoes to be cut covered by each first sub-region and the number of eyes contained in each first sub-region.
[0049] The spliced image is obtained by performing feature matching based on pixel points corresponding to the positioning marks contained in each image in the target image as feature points.
[0050] The step of generating a target cutting scheme in the potato seed potato preparation system provided in the embodiment of the present application may include the following steps 301 to 305:
[0051] Step 301: Place potatoes to be cut on a tray, and control the image acquisition unit to respectively acquire potato images from left and right perspectives, and merge and splice the potato images from the two perspectives to obtain a potato splicing image with a large viewing angle range, so as to display more complete eye information.
[0052] Step 302: Use an improved lightweight target detection network to identify eyeball targets in the target area of the stitched image and generate the eyeball information.
[0053] Step 303: On the stitched image, the potato area is divided into i sub-areas by the center line of the cut groove, and a two-dimensional array is constructed. If the eye detection network (i.e., the improved lightweight target detection network described above) detects an eye in the sub-area to which B[1,i] belongs, then B[1,i] = "True", otherwise B[1,i] = "False". Calculate the potato area of the i-th sub-area and store it in B[2,i].
[0054] Exemplarily, on the stitched image, the target area containing the potatoes to be cut is divided into k×N sub-areas (i.e., the first sub-area mentioned above) by the center line of the cut groove. Construct an M×2 array B to represent these sub-areas. Where M=k×N. Increase the index of array B in a counterclockwise direction, and the first sub-area is calculated from the center line of the tray opening groove: if the eye detection network detects an eye in the area to which B[1,i] belongs, set B[1,i]=“True”, otherwise it is “False”, i=1,…,k×N; calculate the potato area of the i-th sub-area and store it in B[2,i].
[0055] Specifically, the cutting decision module is further configured to generate a plurality of separation lines corresponding to the number and positions of the blades according to the initial positions of the plurality of blades of the tool.
[0056] Specifically, the cutting decision module is further used to rotate the multiple separation lines with the target cutting point as the rotation center to generate multiple cutting plans.
[0057] Among them, one rotation generates a cutting plan, and the angle of each rotation is the same; each segmentation plan can divide the target area into multiple second sub-areas; a second sub-area covers at least one first sub-area; the number of first sub-areas covered by each second sub-area is the same.
[0058] Specifically, the cutting decision module is further used to calculate the area of the potatoes to be cut and the number of eyes in each second sub-region corresponding to each cutting scheme according to the first sub-region covered by each second sub-region corresponding to each cutting scheme; the number of first sub-regions covered by each second sub-region is the same.
[0059] Specifically, the cutting decision module is further configured to calculate the area variance of each cutting scheme based on the area of the potatoes to be cut in each second sub-region corresponding to each cutting scheme.
[0060] Specifically, the cutting decision module is further configured to determine the cutting scheme with the smallest area variance among the multiple cutting schemes and containing buds in each second sub-area as the target cutting scheme.
[0061] It can be understood that, in the embodiment of the present application, each cutting strategy can be regarded as a simulated cutting operation for cutting potatoes, and the optimal cutting strategy is screened out by calculating the cutting result of each simulated cutting operation.
[0062] Step 304: Identify the positioning mark of the tray in the stitched image, determine the point where the center of the cutter falls on the potato using the intersection of the extended lines of the positioning mark, and extract the potato outline from the stitched image using an image segmentation method.
[0063] Exemplarily, N radial lines are drawn radially with the knife landing point as the center. When the knife is in the initial position, the first radial line passes through the cutting groove in the stitched image, and the N radial lines divide the potato area into N separation areas (i.e., the second sub-areas mentioned above) in a counterclockwise direction, simulating the situation where the potato is divided into N pieces when the knife falls.
[0064] Step 305, construct an array C with a dimension of (M-1)×N×2, which stores the simulated tool rotation p times (p=1,…,M-1), the area of the j-th separation area Aj each time it rotates θ, and whether the sub-area covered by Aj (i.e., the first sub-area mentioned above) has a bud eye.
[0065] For example, Figure 3 As shown, after the target area where the potatoes are located in the stitched image is divided into N areas (i.e., the multiple second sub-areas mentioned above), the distribution of the N areas can be adjusted by rotation with the target knife landing point as the rotation center, so as to calculate the detailed information corresponding to each distribution situation.
[0066] In the embodiment of the present application, when N radiation lines are at the initial position, each separation area covers t sub-areas, and C[1, Sj, Yj] is calculated according to the following formula 1 and formula 2:
[0067]
[0068]
[0069] Where Sj is the area of potatoes contained in the jth separation zone, Yj is used to indicate whether the jth separation zone contains buds; B indicates the t sub-areas (i.e., the first sub-area) covered by the jth separation zone (i.e., the second sub-area) when p=1. The ray of the simulated tool rotates by an angle of θ each time. When it rotates to the pth time, C[p,Sj,Yj] is calculated according to the following formulas 3, 4, and 5:
[0070]
[0071]
[0072] q=mod(p-1+(j-1)*t+i,M) (Formula 5)
[0073] Wherein, B is the t sub-areas covered by the j-th separation area when rotated p times.
[0074] In the embodiment of the present application, according to the calculation method of the above array C, the variance of the area Sj of potatoes covered by the N separation zones when the simulated cutter rotates for the pth time is calculated in sequence based on the following formula 6:
[0075]
[0076] in, is the variance of the area of potatoes contained in the N separation zones when the cutter rotates for the pth time, S N It is the area of potatoes covered by the Nth separation zone when the simulated cutter rotates for the pth time.
[0077] After that, according to the above array C and the area variance obtained after each rotation Determine the optimal cutting position. If there is an R-th rotation of the simulated cutter, the variance of the area of potatoes covered by N separation zones is The smallest, and all N separation zones contain buds (ie, Yj = "True"), then the position of the simulated tool's Rth rotation is the optimal cutting position.
[0078] Exemplarily, the driving mechanism is specifically configured to cut the potatoes to be cut on the tray according to the cutting scheme indicated by the target cutting strategy when the position sensor indicates that the tray is located at the second position.
[0079] Specifically, when the position sensor indicates that the tray is located at the second position, a second trigger signal is generated, and the second trigger signal is used to instruct the cutting device to control the driving mechanism to adjust the rotation angle of the tool according to the tool rotation angle indicated by the target cutting strategy, and then control the tool and the tray through the driving mechanism to complete the cutting operation of the potatoes to be cut.
[0080] Exemplarily, the cutting decision module is specifically configured to identify the eyes of the potatoes to be cut from the stitched image based on an improved lightweight target detection network, and generate the eye information.
[0081] Among them, the improved lightweight target detection network is improved based on the YOLOX-nano network; the improved lightweight target detection network deletes some feature layers in the backbone network of the YOLOX-nano network, and adds an attention mechanism module to the spatial pyramid pooling module of the YOLOX-nano network to form an improved backbone network; the improved lightweight target detection network replaces the feature fusion network in the YOLOX-nano network with a deep and shallow fusion feature module as the feature network to realize feature fusion.
[0082] In related technologies, the basic structure of the YOLOX-nano original network consists of the backbone network Backbone, the feature fusion network PAFPN and YOLOX-head. The detection network structure is improved and streamlined with the idea of reducing network depth and improving small target detection accuracy, and a lightweight detection network suitable for bud eyes is designed.
[0083] In the embodiment of the present application, the improved lightweight object detection network deletes the 13×13 feature layer in the Backbone of the original YOLOX-nano network, and adds an attention mechanism module to the Spatial Pyramid Pooling (SPP) module to form an improved backbone network Improve-Backbone. A simple and efficient deep and shallow fusion feature module (DSFM) is used to replace the PAFPN in YOLOX-nano as the feature network to achieve feature fusion.
[0084] Afterwards, the Improve-Backbone, DSFM and YOLOX-head are used to form an improved YOLOX-nano network, and the improved YOLOX-nano network is used to identify the eyes of potatoes in the image.
[0085] In the embodiment of the present application, the input of the deep and shallow fusion feature module MSFD is the feature maps of different scales extracted by Improve-Backbone, where the sizes of the feature maps are 208×208, 104×104, 52×52 and 26×26 respectively. The multi-scale fusion with dual attention (MFDA) algorithm is used to fuse the feature maps of three different scales, and output the predicted feature maps of three sizes of 104×104, 52×52 and 26×26.
[0086] The potato seed potato preparation system provided in the embodiment of the present application, first, the image acquisition device acquires the target image of the potato to be cut placed on the tray. Then, the cutting decision module generates a target cutting strategy according to the target image acquired by the image acquisition device. Finally, the driving mechanism controls the cutter to cut the potato to be cut on the tray according to the cutting scheme indicated by the target cutting strategy, thereby realizing the rapid preparation of potato seed potatoes, and at the same time, each potato tuber of the cut potato contains bud eyes, and the sizes of different potato tubers are similar.
[0087] It should be noted that, in the embodiments of the present application, the potato seed potato preparation system shown in the above-mentioned method drawings is illustrated by taking one of the drawings in the embodiments of the present application as an example. In specific implementation, the potato seed potato preparation system shown in the above-mentioned method drawings can also be implemented in combination with any other drawings that can be combined as shown in the above-mentioned embodiments, which will not be described in detail here.
[0088] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.
[0089] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the system described in each embodiment or some parts of the embodiment.
[0090] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.
Claims
1. A potato seed potato preparation system, characterized in that: The system comprises: a seed cutting device, an image acquisition device and a cutting decision module; the seed cutting device comprises: a knife and a tray, and a driving mechanism for controlling the movement of the knife and the tray; The image acquisition device is used to acquire a target image of the potatoes to be cut placed on the tray; The cutting decision module is used to generate a target cutting strategy according to the target image acquired by the image acquisition device; The driving mechanism is used to control the cutter to cut the potatoes to be cut on the tray according to the cutting scheme indicated by the target cutting strategy, so that each potato tuber after cutting contains bud eyes and different potato tubers are of similar sizes; The tool comprises a plurality of blades fixed on a blade shaft; the angles between any two adjacent blades are the same; the tray is provided with equally spaced cutting grooves matching the blades, and the number of the cutting grooves is an integral multiple of the number of the blades; a circular aperture is provided at the center of the tray, and the axis of the circular aperture is in the same straight line as the blade shaft; a plurality of positioning marks are provided at the edge of the tray; the plurality of positioning marks are used to determine the target landing point of the center of the tool on the potato to be cut; The driving mechanism is specifically used to control the tool and the tray to move toward each other; The tray is also provided with an open slot; the seed potato preparation system further comprises: a tray detachment sensor; The driving mechanism is also used to control the tray to change the movement direction after the seed cutting device completes cutting the potatoes to be cut, so that the knife shaft slides out of the opening groove; The driving mechanism is also used to control the tool and the tray to reset when the tray detachment sensor detects that the knife shaft slides out of the opening groove.
2. The system according to claim 1, characterized in that The image acquisition device includes a plurality of image acquisition units; each of the plurality of image acquisition units is used to acquire images of the potatoes to be cut at different angles; the potato seed potato preparation system also includes: a position sensor; the position sensor is used to detect the position of the tray.
3. The system according to claim 2, characterized in that The image acquisition device is specifically used to control the image acquisition device to acquire the target image when the position sensor indicates that the tray is located at the first position, and send the target image to the cutting decision module; The target image includes images of the potatoes to be cut at different angles acquired by the image acquisition units.
4. The system according to claim 3, characterized in that The cutting decision module is specifically used to generate multiple cutting schemes according to the target image acquired by the image acquisition device, and select a target cutting scheme from the multiple cutting schemes.
5. The system according to claim 4, characterized in that The cutting decision module is specifically used to, after receiving the target image, stitch the images in the target image to generate a stitched image of the maximum viewing angle of the potato to be cut; The cutting decision module is further configured to extract the contour of the potato to be cut in the spliced image, and determine a target area in the spliced image that contains the potato to be cut; The cutting decision module is further configured to generate eye information of the potato to be cut according to the spliced image and the target area; The cutting decision module is further configured to divide the target area into a plurality of first sub-areas along the center of each cutting groove according to the distribution of the plurality of cutting grooves; The cutting decision module is further configured to calculate the area of the potatoes to be cut covered by each first sub-region and the number of eyes contained in each first sub-region; The spliced image is obtained by performing feature matching based on pixel points corresponding to the positioning marks contained in each image in the target image as feature points.
6. The system according to claim 5, characterized in that The cutting decision module is further configured to generate a plurality of separation lines corresponding to the number and positions of the blades according to the initial positions of the plurality of blades of the tool; The cutting decision module is further used to rotate the multiple separation lines with the target cutting point as the rotation center to generate multiple cutting schemes; one rotation generates one cutting scheme, and the angle of each rotation is the same; each segmentation scheme divides the target area into multiple second sub-areas; one second sub-area covers at least one first sub-area; and each second sub-area covers the same number of first sub-areas; The cutting decision module is further configured to calculate the area of the potatoes to be cut and the number of eyes in each second sub-region corresponding to each cutting scheme according to the first sub-region covered by each second sub-region corresponding to each cutting scheme; The cutting decision module is further configured to calculate the area variance of each cutting scheme based on the area of the potatoes to be cut in each second sub-region corresponding to each cutting scheme; The cutting decision module is further specifically used to determine the cutting scheme with the smallest area variance among the multiple cutting schemes and containing bud eyes in each second sub-area as the target cutting scheme.
7. The system according to any one of claims 4 to 6, characterized in that The driving mechanism is specifically configured to cut the potatoes to be cut on the tray according to the cutting scheme indicated by the target cutting strategy when the position sensor indicates that the tray is located at the second position.
8. The system according to claim 5 or 6, characterized in that: The cutting decision is specifically used to identify the eyes of the potatoes to be cut from the spliced image based on an improved lightweight target detection network, and generate the eye information; Among them, the improved lightweight target detection network is improved based on the YOLOX-nano network; the improved lightweight target detection network deletes some feature layers in the backbone network of the YOLOX-nano network, and adds an attention mechanism module to the spatial pyramid pooling module of the YOLOX-nano network to form an improved backbone network; the improved lightweight target detection network replaces the feature fusion network in the YOLOX-nano network with a deep and shallow fusion feature module as the feature network to realize feature fusion.
