Rosa chinensis cut flower grading system and method

Abstract

The invention relates to a Chinese rose cut-flower grading system. The system comprises an image acquisition module for acquiring cut-flower related parameters, a cut-flower transmission module and a control module in communication connection with the image acquisition module and the cut-flower transmission module. The system can adjust the information acquisition content of the image acquisition module based on the adjustment of the sorting standard, and generates a corresponding score for evaluating the cut flower grade based on the control module. And the cut flower conveying module can convey the cut flowers of the corresponding grade to the corresponding sorting area based on the cut flower grade under the control of the instruction sent by the control module. The invention relates to a method for accurately identifying and classifying morphological characteristics of flowers through an imaging technology. Specifically, systematic division is carried out aiming at the flower type, the blooming grade and the deformity degree of the flowers, and seasonal change analysis is carried out on the sizes and the deformity rates of the flowers, so that the accuracy and the reliability of the automatic grading equipment in the flower cutting and grading process are remarkably improved.

Description

Technical Field

[0001] The invention relates to the technical field of automatic flower grading, and in particular to a rose cut flower grading system and method. Background Art

[0002] Yunnan has become the global center of fresh cut flower production. With its resource advantages, the production of fresh cut flowers ranks first in the world. At present, the grading of fresh cut flowers mainly relies on manual grading, which has problems such as low efficiency, lack of precision and inconsistent grading standards, which seriously affects the ornamental quality and economic value of fresh cut flowers.

[0003] The particularity of cut roses lies in the diversity and complexity of their flowers. The subjectivity and sensory ability of manual grading limit the accuracy of grading, resulting in different grading results for the same batch of cut roses due to inconsistent evaluation standards, thus affecting market positioning and selling prices. In some large enterprises and planting bases, although the introduction of automatic grading technology can theoretically improve the objectivity and consistency of grading, the current automatic grading equipment still has the problem of insufficient accuracy in practical application. Especially in identifying deformed flowers and determining the degree of flower opening, existing equipment is difficult to achieve ideal grading effects. The detection accuracy of cut roses not only exists in the selection of quality parameters of cut roses, but also needs to take into account the adjustment of the scoring model for grading cut roses.

[0004] A Chinese patent with publication number CN111495780A discloses a fresh cut flower grading method, which aims to establish a comprehensive evaluation model for the quality of fresh cut flowers by extracting image features of flower damage, flower color, flower opening index, stem length and flower stem inclination, and integrating the grading standards of each image feature. The method provides a basis for extracting flowering index features in the automatic grading of rose cut flowers and determining the storage time of fresh flowers. However, according to the long-term investigation of the rose cut flower market by this application and the feedback from consumers and rose cut flower merchants, the quality grade evaluation of rose cut flowers in the eyes of consumers does not only rely on flower color, flower opening index or stem length, but more on phenotypic traits such as flower size, flower opening degree, and flower deformity that are intuitively distinguished. As a market-oriented commodity, rose cut flowers must fully consider market demand and the market's evaluation of the value of cut flowers. This means that grading is not just a simple judgment of the appearance and quality of flowers, but also a comprehensive analysis of consumer preferences, market trends and competitive environment. Therefore, a reasonable grading method should be able to reflect the actual needs of the market and ensure that each batch of cut flowers can meet the quality standards of the target market and the expectations of consumers. In this way, not only can sales efficiency be improved, but also the optimal allocation of resources can be achieved, ultimately promoting the healthy development of the entire fresh cut flower industry.

[0005] In addition, on the one hand, there are differences in understanding among those skilled in the art; on the other hand, the applicant studied a large number of documents and patents when making the present invention, but due to space limitations, not all details and contents are listed in detail. However, this does not mean that the present invention does not have the characteristics of these prior arts. On the contrary, the present invention already has all the characteristics of the prior art, and the applicant reserves the right to add relevant prior art to the background technology. Summary of the invention

[0006] In order to solve the above technical problems, the present application relates to a cut flower grading system, which includes an image acquisition module for acquiring parameters related to cut flowers, a cut flower conveying module, and a control module that is communicatively connected to the image acquisition module and the cut flower conveying module. The system can adjust the information acquisition content of the image acquisition module based on the adjustment of the sorting standard, and generate a corresponding score for evaluating the grade of the cut flowers based on the control module. Based on the grade of the cut flowers, the cut flower conveying module can convey the cut flowers of the corresponding grade to the corresponding sorting area under the control of the instructions sent by the control module.

[0007] The present application also relates to a rose cut flower grading system, comprising a multi-stage image acquisition module and a control module. The image acquisition module is used to acquire an overall morphological image and / or a local close-up image of the cut flower. The control module is communicatively connected with the image acquisition module.

[0008] According to a preferred embodiment, Figure 3 As shown, the control module is configured as:

[0009] Obtaining a score for flower deformity based on the overall morphological image of the cut flower and the local close-up image of the flower collected by the image acquisition module;

[0010] Obtaining scores for flower size and flower openness based on local close-up images of cut flowers collected by the image acquisition module;

[0011] Obtaining a score for the length of a flower branch based on a local close-up image of a flower stem of a cut flower collected by an image acquisition module;

[0012] The objective weights of flower size, flower openness, branch length and flower deformity were determined according to the weighting method;

[0013] Based on the weighted summation method, the quality score of rose cut flowers is confirmed according to the scores and objective weights of flower size, flower openness, branch length and flower deformity.

[0014] According to a preferred embodiment, the rose cut flower grading system includes a cut flower conveying module, which conveys cut flowers of different grades to different sorting areas based on the cut flower grading evaluation results, wherein the control module in communication with the cut flower conveying module is configured as follows:

[0015] The evaluation result of the cut rose quality score is transmitted to the cut flower transmission module according to the corresponding transmission path of the cut rose.

[0016] According to a preferred embodiment, the parameters of the weighting method include the time that affects the flower deformity rate and flower size. Preferably, the weighting method determines the weights of flower size, flower openness, branch length and flower deformity by: determining the relative weight of each indicator according to the hierarchical analysis method; calculating the information entropy of each indicator in combination with the entropy weight method, and introducing the time factor to evaluate the objective weight of the indicator.

[0017] According to a preferred embodiment, based on the input rose variety, the control module adjusts the objective weights of flower size, flower openness, branch length and flower deformity.

[0018] According to a preferred embodiment, the control module is configured to assign points according to the measured values ​​of flower size, flower openness, flower branch length and flower deformity to quantify flower size, flower openness, flower branch length and flower deformity.

[0019] According to a preferred embodiment, the multi-stage image acquisition module includes a plurality of cameras with different focusing distances and resolutions.

[0020] According to a preferred embodiment, the flower size is the horizontal diameter, the vertical diameter or the ratio of the two of the flower.

[0021] According to a preferred embodiment, flower deformity includes the bend rate and / or the incidence rate of silly flowers.

[0022] The present application also relates to a rose cut flower grading method, comprising the following steps:

[0023] Acquire visible light images of cut flowers;

[0024] Obtain scores for flower size, flower openness, branch length, and flower deformity of cut flowers based on visible light images;

[0025] The objective weights of flower size, flower openness, branch length and flower deformity were determined according to the weighting method;

[0026] The rose cut flower quality score is calculated based on the scores and objective weights of flower size, flower openness, branch length, and flower deformity, among which:

[0027] The parameters of the weighting method include time, which affects the flower deformity rate and flower size. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 This is a structural diagram of a cut flower conveying module shown in Example 1;

[0029] Figure 2This is a structural diagram of the sorting and transporting unit shown in Example 1;

[0030] Figure 3 This is a system information flow chart involved in this application;

[0031] Figure 4 This is a flow chart of the system operation involved in this application;

[0032] Figure 5 This is a schematic diagram of the opening level of the high-angle-shaped "Peach Snow Mountain";

[0033] Figure 6 This is a diagram showing the opening levels of the cup-type 'Naomi';

[0034] Figure 7 This is a schematic diagram of the opening level of the rosette-shaped 'Lotus Goddess';

[0035] Figure 8 This is a schematic diagram of the severity level of the 'Mitaoxueshan' flower.

[0036] Fig. 9 This is a schematic diagram of the severity level of the elbow flower of 'Mitaoxueshan';

[0037] Fig.10 This is a schematic diagram of the change in elbow incidence rate in different flower-type varieties in different seasons;

[0038] Fig.11 This is a schematic diagram of the changes in the incidence of simple-bud flowers in different flower-type varieties in different seasons;

[0039] Fig.12 This is a schematic diagram of the changes in the horizontal diameter of flowers of different flower types and varieties in different seasons;

[0040] Fig.13 This is a schematic diagram of the changes in the longitudinal diameter of flowers of different flower types in different seasons;

[0041] Fig.14 This is a schematic diagram of the changes in the aspect ratio of flowers of different flower types in different seasons;

[0042] Fig.15 This is a structural diagram of another cut flower conveying module shown in the present application, wherein the cut flower conveying module is a FS-ROSE SORTING series of flower automated intelligent grading equipment;

[0043] Fig.16 This is an implementation of the static manual loading unit shown in this application.

[0044] Reference numerals list

[0045] 100: image acquisition module; 200: control module; 300: cut flower conveying module; 310: sorting and conveying unit; 3101: inner plate; 3102: inner transfer chain; 3103: temporary storage tank; 3104: connecting rod; 3105: outer transfer chain; 3106: outer plate; 3107: spiral conveyor belt unloading port; 320: sorting unit; 3201: storage basin; 3202: turntable; 330: sorting and conveying unit; 3301: first conveyor belt; 3302: second conveyor belt; 3303: third conveyor belt; 340: static manual loading unit; 3401: feeding port; 3402: conveyor belt; 3403: sub-bin. DETAILED DESCRIPTION

[0046] The following is a detailed description with reference to the accompanying drawings.

[0047] In the description of the present invention, the terms "first" and "second" are only used for descriptive purposes and are not to be understood as indicating or implying relative importance. In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "provided with", "connected", etc. should be understood in a broad sense. For example, "connected" can be a fixed connection, a detachable connection, or an integral connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal connection of two elements. In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "loose" and "curled" should be understood in a broad sense. For example, it can mean a loose petal structure or a loose petal arrangement. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood based on specific circumstances.

[0048] Determination of flower opening level: The rose cut flower opening level is determined by vase insertion test of the main flower types of the current main cultivated varieties. The flower opening level is divided into 5 levels from level 1 to level 5. Preferably, the flower opening degree is divided according to the opening level. The harvested cut flowers are mainly level 2, 3 and 4. The flower opening degree is divided into large opening degree (wide opening), normal opening degree (medium opening) and small opening degree (small opening). Figures 5 to 7 The flower morphology of the high-heart and upturned-angled 'Peach Snow Mountain', the cup-shaped 'Naomi' and the rosette-shaped 'Lotus Goddess' at different opening levels is shown.

[0049] Openness level 1: Petals are tightly packed and sepals have already unfolded. Openness level 2: Outer petals begin to loosen, inner petals are tightly packed. Openness level 3: Outer petals are loosely unfolded, inner petals begin to loosen. Openness level 4: Outer petals are curled outwards, and multiple layers of petals are unfolded. Openness level 5: Petals are loose all over, and multiple layers of petals are curled.

[0050] Flower deformity includes elbow rate and / or incidence of silly bud flowers. Silly bud flowers mean that the buds of rose cut flowers are tightly wrapped and the flowers cannot open normally. Elbow flowers refer to the deformed state of flowers in which the flower stems under the flower heads of rose cut flowers are bent. The degree of opening of silly bud flowers and the angle of elbow flowers are divided into mild, moderate and severe. In this application, mild silly bud flowers have 3 to 4 layers of petals on the outer layer unfolded, and the inner petals are slightly loose; moderate silly bud flowers have 1 to 2 layers of petals on the outer layer unfolded, and the inner petals are tightly wrapped; severe silly bud flowers have less than 1 layer of petals on the outer layer unfolded, and the inner petals are tightly wrapped. The elbow angle of mild elbow flowers is less than 30 degrees, the elbow angle of moderate elbow flowers is between 30 and 60 degrees, and the angle of severe elbow flowers is greater than 60 degrees. Figure 8 The flower morphology of 'Mitaoxueshan' at different flower grades is shown. Fig. 9 The flower morphology of 'Mitaoxueshan' at different elbow grades is shown.

[0051] The flower size is the horizontal diameter, vertical diameter or the ratio of the two. Taking the horizontal diameter and vertical diameter of the flower as an example, the horizontal diameter of the small flower is less than 40 mm and the vertical diameter is less than 40 mm; the horizontal diameter of the medium flower is between 40 and 70 mm and the vertical diameter is between 40 and 60 mm; the horizontal diameter of the large flower is greater than 70 mm and the vertical diameter is greater than 60 mm.

[0052] The length of the flower branch refers to the length of the branch at the base of the flower branch. According to the length of the flower branch, it is divided into long branches, medium branches and short branches. The length of short branches is less than 40 cm, the length of medium branches is 40~70 cm, and the length of long branches is greater than 70 cm.

[0053] Regarding the classification of rose cut flower shapes, the current common rose cut flower shapes mainly include high heart shape, cup shape, rosette shape, flat open shape, ball shape, curled shape, wavy shape or button eye shape.

[0054] The high-heart type includes the high-heart angled type and the high-heart notched type. The high-heart angled type has a high and tightly wrapped flower core, and is also characterized by the outer petals turning up. During the development of the high-heart notched type flower, when the flower core does not appear, the flower shows two centers.

[0055] The cup type includes deep cup type, shallow cup type and open cup type. The petals of the deep cup type grow upward, presenting a more three-dimensional cup-shaped flower shape as a whole. The petals of the shallow cup type grow upward, presenting a shallow cup flower shape between the flat open shape and the deep cup shape as a whole. The petals of the open cup type grow upward and move closer to the outer petals, thus forming a hollow cup-shaped flower shape with the stamens exposed.

[0056] Rosette type includes rosette type and quarter rosette type. Rosette type flowers are composed of dense petals, with wide outer petals and small inner petals tightly packed together, presenting a rosette shape as a whole. Quarter rosette type petals have obvious groupings (not necessarily four groups) inside, presenting a rosette shape.

[0057] Flat-opening types include single petals and semi-double petals. The petals of flat-opening flowers grow shallowly and flatly around them, giving the overall flower a flat shape.

[0058] Ball-shaped flowers include ball-shaped and pompom-shaped flowers. Ball-shaped petals surround the center of the inflorescence, hiding the stamens, almost forming a round ball. Pompom-shaped flowers are small, round, and composed of tiny petals.

[0059] The curling type is a flower shape in which the petals are curled up as a whole, causing the flower to curl outwards.

[0060] The wavy type has wavy edges on the petals.

[0061] The inner petals in the center of the button-eye flower are turned upside down to form a button eye.

[0062] The most common rose cut flower varieties currently grown in the market have flower shapes mainly of high heart, cup and rosette types.

[0063] Weighting method: By assigning weights to multiple characteristic parameters, the overall quality of cut flowers can be objectively evaluated. In this technical solution, the weighting method means that multiple evaluation indicators of rose cut flowers are weighted to achieve a comprehensive score of flower quality. The basis of the weighting method is multi-criteria decision making (MCDM), which can effectively handle complex decision-making problems involving multiple evaluation criteria. By objectively scoring and assigning weights to each evaluation indicator, a systematic and quantitative evaluation system is formed.

[0064] Example 1

[0065] This embodiment provides a rose cut flower grading system for grading cut flowers based on cut flower grades. This embodiment provides a cut flower conveying module 300. This embodiment provides a cut flower conveying module 300 for grading.

[0066] The rose cut flower grading system comprises a control module 200 and a cut flower conveying module 300 controlled by the control module 200 .

[0067] The cut flower conveying module 300 is provided with a sorting conveying unit 310 for grade evaluation, a sorting unit 320 and a plurality of sorting conveying units 330 .

[0068] like Figure 1As shown, the sorting and transporting unit 310 can be set as a spiral conveyor belt. The spiral conveyor belt is a three-dimensional conveyor belt that is surrounded along the central axis and extends in the height direction. The spiral conveyor belt is provided with multiple layers of ring belts in the height direction, so that the cut flowers can be displaced in the height direction under the action of the spiral conveyor belt. Specifically, an inner plate 3101 is provided inside the spiral conveyor belt. Inside the inner plate 3101, the inner transport chain 3102 is combined with the temporary storage tanks 3103 of the upper and lower layers to ensure that the cut flowers can be smoothly transferred from the conveyor belt and temporarily stored. The outer transport chain 3105 is located on the outside of the sorting and transporting unit 310. The outer plate 3106 surrounds the entire structure, plays a role of protection and support, and ensures the stability and safety of the cut flowers during the transmission process. The outer transport chain 3105 is arranged inside the outer plate 3106. The outer transport chain 3105 and the inner transport chain 3102 are respectively connected to the connecting rod 3104. The connecting rod 3104 passes through the temporary storage tank 3103, so that the temporary storage tank 3103 can move synchronously with the connecting rod 3104 controlled by the transfer chain. At the unloading port 3107 of the spiral conveyor, through the coordinated action of the inner transfer chain 3102 and the outer transfer chain 3105, the upper temporary storage tank 3103 completes the flipping action and turns into the lower temporary storage tank 3103. More specifically, as Figure 2 As shown, at the position of the spiral conveyor belt discharge port 3107, the upper temporary storage tank 3103 with an opening facing upward is further turned into a temporary storage tank 3103 with an opening facing the spiral conveyor belt discharge port 3107. Then, the temporary storage tank 3103 with an opening facing the spiral conveyor belt discharge port 3107 is turned into a lower temporary storage tank 3103 with an opening facing downward.

[0069] Based on the cut flower grading evaluation results, the cut flowers of different grades are transported to different sorting areas. Figure 1 As shown, the sorting unit 320 is arranged at the position of the spiral conveyor discharge port 3107 of the sorting and transporting unit 310. The sorting unit 320 comprises a storage basin 3201 and a turntable 3202. The turntable 3202 can be controlled by the control module 200 to adjust the rotation angle. A plurality of storage basins 3201 are arranged on the turntable 3202. Different storage basins 3201 store cut flowers of different grades. For example: when the grade of cut flowers stored in the temporary storage slot 3103 closest to the spiral conveyor belt unloading port 3107 is level one, the control module 200 sends an instruction including a rotation angle to the turntable 3202, and the turntable 3202 can make the storage basin 3201 for storing level one cut flowers arranged thereon closest to the spiral conveyor belt unloading port 3107, so that when the temporary storage slot 3103 for temporarily storing the cut flowers is flipped to be the temporary storage slot 3103 with its opening facing the spiral conveyor belt unloading port 3107, the cut flowers can fall into the corresponding storage basin 3201.

[0070] like Figure 1As shown, the number of the sorting and transporting units 330 is consistent with the number of the cut flower grades. For example, in this embodiment, the cut flower grades include three grades, namely, grade one, grade two, and grade three, and the sorting and transporting unit 330 includes a first conveyor belt 3301, a second conveyor belt 3302, and a third conveyor belt 3303. Relying on manual or other sorting equipment (such as an industrial robot), the cut flowers in the storage basin 3201 containing the corresponding grades are sorted to the corresponding first conveyor belt 3301, the second conveyor belt 3302, and the third conveyor belt 3303.

[0071] The control module 200 in communication connection with the cut flower delivery module 300 is configured to transmit the corresponding rose cut flower transmission path to the cut flower delivery module 300 based on the evaluation result of the rose cut flower quality score. Preferably, each temporary storage slot 3103 corresponds to a characteristic mark. When the image detection is completed, the mark of the temporary storage slot 3103 can be associated with the information related to its quality score. At the same time, the processor associates the transmission path of the mark. For example: a corresponding information set is generated with the mark. Preferably, the transmission path includes the information of the corresponding storage basin 3201.

[0072] According to a preferred embodiment, the control module 200 includes a data processing unit, a data storage unit, a communication component and a power management component. The data processing unit can be a single-chip microcomputer, a higher performance microprocessor, a multi-core CPU or a dedicated image processing chip, which can process data streams in real time. The data storage unit is, for example, a configuration of an SD card, a high-speed solid-state hard disk or an external flash memory, which is used to store the collected image data as well as the intermediate data and the final result during the processing. The power management component is used to provide a stable power supply for the entire system, such as a DC-DC converter or a battery-powered system. The communication component includes a variety of interfaces, such as USB, Wi-Fi, Bluetooth or Ethernet interfaces, which are used to support data transmission and remote control, facilitating the interaction between the user and the system.

[0073] The image acquisition module 100 is used to acquire the overall morphological image and / or local close-up image of the cut flower. The image acquisition module 100 includes a camera and a support structure for placing the camera. The support structure is, for example, a door frame bracket. The door frame bracket includes a crossbeam and two parallel support rods respectively connected to the crossbeam. The camera is connected to the crossbeam through a rotation unit, and the camera angle and range are adjusted under the control of the control module 200. The other ends of the two support rods connected to the crossbeam are respectively arranged on both sides of the assembly line of the sorting and transmission unit 310. Preferably, the relative position of the image acquisition module 100 and the assembly line of the sorting and transmission unit 310 is fixed. Based on the flow of the assembly line, the image acquisition module 100 can acquire the cut flower image of the area transmitted to the image acquisition range in real time. For example: Figure 1As shown, the two supporting rods of the image acquisition module 100 are respectively connected to the outer plate 3106 and the inner plate 3101 of the sorting and transporting unit 310. The camera of the image acquisition module 100 is arranged above the sorting and transporting unit 310.

[0074] According to a preferred embodiment, the multi-stage image acquisition module 100 includes multiple cameras with different focusing distances and resolutions. For the local close-up image of the cut flower, the camera is, for example, a high-resolution CMOS or CCD image sensor. For the overall morphological image of the cut flower, the camera is, for example, a USB network camera. Based on the difference in focusing points, the camera should also be divided into a macro lens for collecting local close-up images of the cut flower and a standard lens for collecting the overall morphological image of the cut flower.

[0075] According to a preferred embodiment, the cut flower conveying module 300 can also be as follows: Fig.15 The flower automatic intelligent grading device shown. The image acquisition module 100 carried in the flower automatic intelligent grading device includes an industrial computer (such as an industrial-grade image processing computer), a graphics card (such as an AI image processing graphics card), and a synchronization box (such as a signal controller). The flower automatic intelligent grading device includes a static manual feeding unit 340 and a transmission line. Preferably, the image acquisition module 100 is integrated with the control module 200.

[0076] like Fig.16 As shown, the static manual loading unit 340 is a material conveying and loading device. Preferably, the static manual loading unit 340 is configured as a flower unloading component. The flower unloading component includes an inlet 3401, a conveyor belt 3402 and a sub-bin 3403. The sub-bin 3403 is provided with a plurality of outlets based on the grade of the cut flowers. For example, there are three grades of cut rose flowers, and inlets 3401 which are multiples of 3 are provided, and each outlet corresponds to one grade. The sub-bin 3403 is a conveying structure provided with a gentle slope. The area of ​​the conveyor belt 3402 close to the inlet 3401 is a flat area, which is provided with an image acquisition module 100. Preferably, as Fig.16 As shown, based on the characteristic that the diameter of cut flowers is larger than the diameter of flower stems, the static manual feeding unit 340 is configured as a hollow structure (eg, ring-shaped), the flower stems pass through the hollow area, and the flowers are engaged on both sides.

[0077] When the cut flowers enter from the feed port 3401, the image acquisition module 100 can collect relevant image data. Based on the instructions sent by the control module 200, the conveyor belt 3402 conveys the cut flowers to the discharge port of the corresponding level of the bin 3403, and slides from the discharge port of the corresponding level to the transmission line based on the height difference. The static manual feeding unit 340 is connected to the transmission line, and the cut flowers are conveyed from the static manual feeding unit 340 to the transmission line, and then the cut flowers that have been classified are packaged manually.

[0078] Specifically, Fig.15 , 16 The arrangement of the cut flower conveying module 300, the control module 200 and the image acquisition module 100, and the actual operation method thereof comprises the following steps:

[0079] Flowers are loaded on the static manual loading unit 340, and the image acquisition module 100 acquires image information of the cut flowers;

[0080] Based on the image information collected by the image collection module 100, the control module 200 generates information on the grade of the cut flowers (or generates information on the discharge port to which the cut flowers should be delivered), and sends it to the cut flower delivery module 300;

[0081] The conveyor belt 3402 conveys the cut flowers to the corresponding discharge port, so that the cut flowers can be moved from the discharge port of the corresponding level to the transmission line.

[0082] Example 2

[0083] The hardware configuration of this embodiment is consistent with that of the above-mentioned embodiment, and only improvements are made in data collection and processing.

[0084] The control module 200 is configured to:

[0085] Image acquisition: control the image detection module to acquire images and set the acquisition frequency, resolution and other parameters;

[0086] Image processing: pre-processing the collected images, including denoising, enhancement, cropping, and extracting key features of flowers, such as color, shape, and texture;

[0087] Feature analysis: using machine learning algorithms or deep learning models to analyze the extracted features and identify the types of flowers;

[0088] The grading calculation, based on the preset algorithm and weight ratio, comprehensively considers various parameters of the flowers, including flower size, degree of flower openness, branch length, and flower deformity, and gives the final grading result.

[0089] Specifically, Figure 3 As shown, the control module 200 is configured as follows:

[0090] Obtaining a score of flower deformity based on the overall morphological image of the cut flower and the local close-up image of the flower collected by the image acquisition module 100;

[0091] Obtaining scores for flower size and flower openness based on the local close-up images of cut flowers collected by the image collection module 100;

[0092] The score of the flower branch length is obtained based on the local close-up image of the flower stem of the cut flower collected by the image collection module 100.

[0093] This embodiment provides the following definitions:

[0094] For flower size, large, medium, and small are assigned values ​​of 3, 2, and 1 respectively. For flower openness, large, medium, and small are assigned values ​​of 3, 2, and 1 respectively. For flower branch length, long branches, medium branches, and short branches are assigned values ​​of 3, 2, and 1 respectively. For flower deformity, the mild value of the simple-bud flower and the elbow flower is assigned a value of -1, the moderate value is assigned a value of -2, and the severe value is assigned a value of -3. Score flower size 3, 2, 1; Score the flower openness as 3, 2, or 1; Score the length of the flower branch as 3, 2, or 1; Score -1, -2, -3 for flower deformities.

[0095] like Figure 4 As shown, the objective weights of flower size, flower openness, branch length and flower deformity were determined according to the weighting method.

[0096] like Figure 4 As shown, based on the weighted summation method, the quality score of rose cut flowers is determined according to the scores and objective weights of the flower size, flower openness, branch length and flower deformity. The calculation formula is shown in formula (1):

[0097] … (1)

[0098] Among them, W 大小 Indicates the weight value of the flower size; W 开放度 Indicates the weight value of flower opening degree; W 长度 Indicates the weight value of the branch length; W 畸形 Indicates the weight value of flower deformity.

[0099] This embodiment provides an example. In this example, the weights are: ; ; ; If a flower is medium-sized, medium-open, medium-branched, and slightly budded, then ; ; ; .

[0100] The final total score of the flowers in this example :

[0101]

[0102]

[0103] According to the grade corresponding to the score, the control module 200 controls the sorting unit 320 to transfer the cut flowers transmitted from the sorting and transporting unit 310 to the sorting and transporting unit 330 of the corresponding grade.

[0104] Example 3

[0105] The hardware configuration of this embodiment is consistent with that of the previous embodiment, and only improvements are made in data collection and processing. In this embodiment, based on the seasonal changes in flower size and flower deformity rate, the weight ratio of flower size and flower deformity rate needs to be adjusted accordingly according to the rose species and month.

[0106] Figures 10 to 14 Corresponding to the data in Tables 1 to 5 respectively.

[0107] like Fig.10 As shown in Table 1, the weight ratio is related to time (month). By randomly selecting 300 different rose cut flowers in different months, it is found that the elbow rate of rose cut flowers of three varieties, 'Peach Snow Mountain', 'Naomi' and 'Lotus', showed an approximate change trend in different months, in the same greenhouse (or environment) and under the same culture conditions (the statistical data shown in the table are all rounded to single digit results). Since February, the elbow rate has gradually decreased until July. After July, the elbow rate began to increase. At the same time, the change range of roses of different varieties varies in the same month, among which the change in the elbow rate from July to August is particularly obvious. This result shows that rose variety is also an important influencing factor affecting the change in the elbow rate amplitude. Therefore, the weight adjustment method of different varieties in each month in this embodiment is different.

[0108] Table 1

[0109]

[0110] Similarly, by counting the number of buds of ‘Mitao Snow Mountain’, ‘Naomi’ and ‘Luo Shen’, as well as the horizontal and vertical diameters of the flowers of ‘Mitao Snow Mountain’, ‘Naomi’ and ‘Luo Shen’, as shown in Tables 2 to 4 and Figure 10~Figure 14 , and found that the buds and flower sizes of roses are also affected by the month.

[0111] Table 2

[0112]

[0113] According to Table 2 and Fig.11, the rose cut flowers of the 'Lotus' variety basically do not have simple bud flowers, so the weight ratio of simple bud flowers in its comprehensive score is a fixed value. The number of simple bud flowers of the rose cut flowers of the Mitaoxueshan variety is higher, so the weight of simple bud flowers of the rose cut flowers of the Mitaoxueshan variety is higher.

[0114] Table 3

[0115]

[0116] like Fig.12 As shown in Table 3, the transverse diameter of the 'Peach Snow Mountain' rose cut flower increased slightly from about 59 mm to about 64 mm between January and March; the transverse diameter was relatively stable between April and December, fluctuating between 54 mm and 64 mm, with a downward trend in some months.

[0117] The horizontal diameter of the cut rose 'Naomi' slightly increased from about 58 mm to 60 mm between January and March, and gradually decreased from 60 mm to about 52 mm between April and September, until it rose to 60 mm in October and stabilized within 63 mm in November and December. The results show that the horizontal diameter of 'Naomi' slightly increased at the beginning of the year, then gradually decreased, and slightly recovered at the end of the year.

[0118] The transverse diameter of the rose cut flowers of 'Lotus Goddess' rose increased significantly from January to March, from about 62 mm to 70 mm; from March to June, it remained stable at 69-70 mm; from July to September, the transverse diameter gradually decreased from 67 mm to about 57 mm; from October to November, the transverse diameter increased significantly to a maximum of 70 mm; and decreased to 63 mm in December. The results showed that the transverse diameter of 'Lotus Goddess' changed most significantly, with a large increase at the beginning and end of the year, and fluctuations in the middle period.

[0119] Table 4

[0120]

[0121] like Fig.13 As shown in Table 4, the longitudinal diameter of the cut rose of ‘Mi Tao Xue Shan’ gradually increased from January to June, reaching about 58 mm; from July to September, the longitudinal diameter gradually decreased to 49 mm; from October to December, the longitudinal diameter showed obvious fluctuations, ranging from 55 to 49 mm. The results show that the longitudinal diameter of ‘Mi Tao Xue Shan’ has an upward trend at the beginning and end of the year, and gradually decreases in the middle period.

[0122] The vertical diameter of the cut rose 'Naomi' slightly decreased from about 56 mm to 55 mm between January and February; it increased to 60 mm between February and March; it gradually decreased from 60 mm to about 49 mm between March and September; it slightly increased to 57 mm between September and December. The results show that the vertical diameter of 'Naomi' decreased in the first half of the year, and then gradually increased in the second half of the year.

[0123] The longitudinal diameter of the cut rose of 'Lotus Goddess' rose increased significantly from January to March, from about 57 mm to 61 mm; from March to August, the longitudinal diameter fluctuated and basically stabilized between 59 and 61 mm; from September onwards, the longitudinal diameter decreased from 61 mm to a minimum of 55 mm. The results show that the longitudinal diameter of 'Lotus Goddess' changed most significantly, with large changes at the beginning and end of the year, and remained stable in the middle period.

[0124] Table 5

[0125]

[0126] like Fig.14 As shown in Table 5, the ratio of 'Mi Tao Xue Shan' rose roses from January to March, reaching a peak; it drops significantly to 1.11 in April, and then gradually rises until August; it gradually decreases from August to December. The results show that the 'Mi Tao Xue Shan' rose fluctuates significantly, and the ratio will repeat in multiple time periods.

[0127] The ratio of horizontal diameter to vertical diameter of 'Naomi' rose cut flowers was stable from January to April, ranging from 1.04 to 1.06; it rose to the first peak in May, with a peak of 1.1, and then slightly decreased until November; the ratio increased from November to December. The results show that the ratio of horizontal diameter to vertical diameter of 'Naomi' changed little, and the overall trend was relatively stable, but there was an upward trend at the end of the year.

[0128] The ratio of horizontal diameter to vertical diameter of rose cut flowers of 'Lotus Goddess' rose ...

[0129] Based on the above statistical results, this example proposes a weight adjustment model suitable for different varieties of cut roses.

[0130] The weight coefficient can be adjusted every month based on time (month) as a reference factor. Please refer to formula (2).

[0131] … (2)

[0132] in, is the benchmark weight; is the weight of each month; It is a variety In the month Elbow rate; It is the benchmark elbow rate of the variety (it can be the elbow rate of a previous month, or the calculated average value).

[0133] Take the change of elbow rate of 'Peach Snow Mountain' as an example:

[0134] The bending rate of 'Peach Snow Mountain' was the highest in August, September and October, reaching 1.2%. The bending rate of 'Peach Snow Mountain' was the lowest in July, at 0.4%.

[0135] The elbow rate of 'Mitao Snow Mountain' has the following changing trends: it gradually decreases in winter (January to April), remains low in spring (May to July), rises sharply in summer (August to September), and decreases slightly in autumn (October to December).

[0136] Benchmark weight of elbow rate of 'Peach Snow Mountain' , base elbow rate (Average of elbow rate in the 12 months of the previous year); elbow rate in April of the previous year , then the weight of the April peach snow mountain elbow rate is 0.25.

[0137] Based on long-term market statistics and expert analysis of different flower types of rose cut flowers, the present application finds that the flower size, flower openness, peduncle length or flower deformity of different flower types have a certain impact on consumers' aesthetics. For example: the beauty of high heart-shaped flowers is very dependent on their symmetry and shape, and deformity will significantly reduce their aesthetic value, and high heart-shaped flowers have a higher incidence of simple buds than other flower types. Rosette-shaped flowers generally have larger petals, and the size of the flower directly affects its visual impact and attractiveness. Consumers are more inclined to choose large and full rosette-shaped flowers, so the size of the flower has a more obvious impact on the aesthetic tendency of rosette-shaped flowers.

[0138] According to a preferred embodiment, the system involved in the present application can adjust the weight ratio of flower size, flower openness, flower stem length, and flower deformity based on the flower shape, as shown in Table 6.

[0139] Table 6

[0140]

[0141] Example 4

[0142] The hardware configuration of this embodiment is consistent with that of the above-mentioned embodiment, and only improvements are made in data collection and processing.

[0143] This application proposes a method for adjusting the weight of each indicator in the scoring based on the statistics of rose indicator data in the previous year. Specifically, the weight of flower size, flower deformity, branch length or flower openness is adjusted by combining the analytic hierarchy process (AHP) and the entropy weight method, and introducing the time factor.

[0144] First, the importance of flower size, flower deformity, branch length, or flower openness was scored based on multiple aspects, and the relative weight of each indicator was calculated using the following method: the weight vector was calculated using a pairwise comparison matrix, a consistency test was calculated, and the relative weight of the indicator was confirmed when the consistency of the matrix was acceptable.

[0145] Specifically, the pairwise comparison matrix is ​​shown in formulas (3), (4) and (5):

[0146] … (4)

[0147] … (5)

[0148] in, is a judgment matrix; W is a weight vector; n represents the total number of indicators, that is, the number of rows and columns in the pairwise comparison matrix (preferably, in this embodiment, n is 4); i represents the row index in the pairwise comparison matrix, which usually corresponds to a specific indicator, such as flower size, flower openness, etc.; j represents the column index in the pairwise comparison matrix, which usually refers to other indicators compared with the i-th indicator, for example: if j=2, it means the second indicator (such as flower openness) compared with the i-th indicator (flower size); It is The indicator and The relative importance score of each indicator; represents the normalized weight matrix; A ij It is The indicator and The judgment matrix of the indicators.

[0149] Example: To construct a pairwise comparison matrix A, this embodiment provides evaluation data based on the importance of each indicator by experts. The scoring uses a scale of 1 to 9. 1 means that both are equally important; 3 means that one is slightly important; 5 means that one is important; 7 means that one is very important; 9 means that one is extremely important (the pairwise comparison matrix is ​​symmetrical, if the importance of A relative to B is evaluated as x, then the importance of B relative to A is 1 / x). See the calculation process of the following formula:

[0150]

[0151]

[0152]

[0153] According to a preferred embodiment, a consistency check is performed to confirm the consistency of the matrix. The consistency check formula is shown in formulas (6) to (8).

[0154] … (6) … (7)

[0155] Among them, λ max Indicates the maximum eigenvalue of the matrix; CI (Consistency Index) is a consistency index used to measure the consistency of the matrix; CR (Consistency Ratio) is a consistency ratio used to judge whether the consistency of the matrix is ​​within an acceptable range; n represents the order of the matrix; RI is a random consistency index (Random Index), and its value depends on the order of the matrix. Generally speaking, when n=1, RI is 0.00; when n=2, RI=0.00; when n=3, RI=0.58; when n=4, RI=0.90; when n=5, RI=1.12.

[0156] If CR is less than 0.1, the consistency of the judgment matrix A is acceptable.

[0157] After obtaining the relative weight (i.e., weight vector W), the entropy weight method calculates the objective weight The calculation formulas are shown in formulas (9) and (11).

[0158] … (9) … (10) … (11)

[0159] in, P ij For the Indicator No. The proportion of samples; is a normalization constant, usually taken as ; m is the sample size.

[0160] Here is an example:

[0161] The number of simple buds per 100 flowers of Mitao Snow Mountain in each month in the past year was 24, 16, 12, 10, 8, 7, 6, 5, 4, 2, 6, and 18. The elbow rate (%) of Mitao Snow Mountain in each month in the past year was 0.8, 0.7, 0.6, 0.5, 0.6, 0.5, 0.4, 1.2, 1.2, 1.2, 1.1, and 0.9. The relative weight W of flower deformity in Mitao Snow Mountain is 0.2.

[0162] According to formula (9), the data is normalized and calculated: , , get the normalized value for each month.

[0163] In this example, m is 12, so k is 0.402.

[0164] is 0.918; is 0.920, objective weight , then the objective weight of bud deformity It is 0.2+0.081=0.281.

[0165] It should be noted that the above-mentioned specific embodiments are exemplary, and those skilled in the art can come up with various solutions inspired by the disclosure of the present invention, and these solutions also belong to the disclosure scope of the present invention and fall within the protection scope of the present invention. Those skilled in the art should understand that the present invention specification and its drawings are illustrative and do not constitute a limitation of the claims. The scope of protection of the present invention is defined by the claims and their equivalents. The present invention specification contains multiple inventive concepts, such as "preferably" and "according to a preferred embodiment", both of which indicate that the corresponding paragraph discloses an independent concept, and the applicant reserves the right to file a divisional application based on each inventive concept. Throughout the text, the features guided by "preferably" are only an optional method and should not be understood as being required. Therefore, the applicant reserves the right to abandon or delete the relevant preferred features at any time.

Claims

1. A rose cut flower grading system, characterized in that: Include A multi-stage image acquisition module (100) for acquiring an overall morphological image and / or a local close-up image of a cut flower; and A control module (200) communicatively connected to the image acquisition module (100), wherein the control module (200) is configured as follows: Obtaining a score for flower deformity based on the overall morphological image of the cut flower and the local close-up image of the flower collected by the image collection module (100); Obtaining scores for flower size and flower openness based on the local close-up images of cut flowers collected by the image collection module (100); Obtaining a score for the length of a flower branch based on a local close-up image of a flower stem of a cut flower collected by the image collection module (100); The objective weights of flower size, flower openness, branch length and flower deformity were determined according to the weighting method; Based on the weighted summation method, the quality score of rose cut flowers is confirmed according to the scores and objective weights of the flower size, flower openness, branch length and flower deformity.

2. The rose cut flower grading system according to claim 1, characterized in that: The rose cut flower grading system comprises a cut flower conveying module (300) which conveys cut flowers of different grades to different sorting areas based on the cut flower grading evaluation results, wherein the control module (200) in communication with the cut flower conveying module (300) is configured as follows: Based on the evaluation result of the rose cut flower quality score, the corresponding rose cut flower transmission path is transmitted to the cut flower transmission module (300).

3. The rose cut flower grading system according to claim 1, characterized in that: The parameters of the weighting method include time which affects the flower deformity rate and flower size.

4. The rose cut flower grading system according to claim 3, characterized in that: The method for determining the weights of flower size, flower openness, branch length and flower deformity by the weighting method is as follows: determining the relative weight of each indicator according to the hierarchical analysis method; calculating the information entropy of each indicator in combination with the entropy weight method, and introducing the time factor to evaluate the objective weight of the indicator.

5. The rose cut flower grading system according to claim 4, characterized in that: Based on the input rose variety, the control module (200) adjusts the objective weights of the flower size, flower openness, flower branch length and flower deformity.

6. The rose cut flower grading system according to claim 1, characterized in that: The control module (200) is configured to assign points according to the measured values ​​of flower size, flower openness, flower branch length and flower deformity to quantify the flower size, flower openness, flower branch length and flower deformity.

7. The rose cut flower grading system according to claim 1, characterized in that: The multi-stage image acquisition module (100) comprises a plurality of cameras with different focusing distances and resolutions.

8. The rose cut flower grading system according to claim 1, characterized in that: The flower size is the horizontal diameter, vertical diameter or the ratio of the two of the flower.

9. The rose cut flower grading system according to claim 1, characterized in that: The flower deformity includes the incidence of elbow and / or bud flowers.

10. A method for grading cut roses, characterized in that: The following steps are involved: Acquire visible light images of cut flowers; Obtain scores for flower size, flower openness, branch length, and flower deformity of cut flowers based on visible light images; The objective weights of flower size, flower openness, branch length and flower deformity were determined according to the weighting method; The rose cut flower quality score is calculated based on the scores and objective weights of flower size, flower openness, branch length, and flower deformity, among which: The parameters of the weighting method include time which affects the flower deformity rate and flower size.

Images