A method and system for predicting embryo sac parameters of fresh tangerine peel

Abstract

The invention relates to a method and system for predicting parameters of embryo sac of fresh tangerine peel fruit. The method adopts binocular vision to measure the volume of fresh tangerine peel fruit and correct it. A weighing module with a cantilever beam structure is used to amplify and process a weight signal to obtain a weight parameter. The volume of the tangerine peel fresh fruit and the volume of the embryo sac of the fresh tangerine peel fruit are calculated according to density information. A neural network model is used to process data to obtain accurate embryo sac parameters of fresh tangerine peel fruit. The method is convenient for generating optimized processing tool feeding parameters and tool expansion parameters according to the prediction results of the embryo sac of fresh tangerine peel fruit, controlling the automatic operation of rotary digging and removing the embryo sac, and improving the effect and speed of removing the sac.

Description

Technical Field

[0001] The invention relates to the field of Chinese medicinal material processing, and in particular to a method and system for predicting embryo sac parameters of fresh tangerine peel fruit. Background Art

[0002] Huajuhong is a precious Chinese medicinal material produced in Huazhou, Guangdong. It is based on the original plant Huazhou pomelo. The peel of the fresh Huajuhong fruit is cut into five or seven petals and the pulp is removed. The peel is processed and dried by traditional technology to become the original five-claw and seven-claw Huajuhong medicinal materials. It has the effects of invigorating the stomach and promoting qi, relieving cough and reducing phlegm. Therefore, the decapsulation of fresh Huajuhong fruit is an indispensable step in the traditional Huajuhong processing. In recent years, the fetal fruit with a fruit age of 41-55 days is mostly used as the original hairy juhong medicinal material, and it does not need to be decapsulated. The original fruit with a fruit age of more than 55 days and a fruit diameter of more than 90mm generally needs to be decapsulated. At present, the fresh Huajuhong fruit is mainly picked by hand, and the suitable picking period is very short. With the continuous expansion of the planting area of ​​Huajuhong, as a seasonal temporary worker, there is a shortage of picking labor and insufficient picking capacity. The fruits that are not picked in time grow rapidly and need to be decapsulated. Half of the fresh fruits picked each year need to be decapsulated. At present, the removal of pods from fresh tangerine peel mainly relies on manual cutting of the peel and digging out of the pulp, which is slow and costly. In addition, the shortage of labor has limited processing capacity, restricting the company's benefits and development.

[0003] The prior art provides a fresh fruit decapsulation device that can control the feed depth and expansion size of the tool by obtaining or setting parameters, so as to better decapsulate the orange peel fresh fruits with different fruit stems and embryo sac sizes; however, in actual agricultural product processing, especially as individual orange peel fresh fruits, the difference is relatively large from the early spherical shape to the ellipsoidal shape or the grapefruit shape with obvious head size. After the fruit age is more than 55 days, the overall volume increase rate of the fruit is much lower than the internal embryo sac increase rate, the peel and the orange white part gradually become thinner and the embryo sac gradually increases, and there are new changes every day. Due to the differences in flowering and pollination time and growth differences of individual fruit trees, even the orange peel fresh fruits of the same fruit tree have large individual differences, which may cause the orange peel fresh fruits picked on the same day to have similar external volume and fruit shape, and the internal embryo sac size may also be quite different, and the orange peel fresh fruits picked on different days are even more different. Therefore, relying on artificial experience to set a processing tool feed parameter and tool expansion parameter with strong universality requires high experience and poor universality, and it is difficult to find a suitable parameter suitable for batch processing of orange peel decapsulation. There are also obvious differences in the properties of the tangerine peel, tangerine white and embryo sac. After cutting, the size of the embryo sac can be distinguished and evaluated with the help of machine vision and other technologies. However, the fresh tangerine fruit needs to be cut in half for visual detection and milling of the embryo sac, which destroys the shape of the tangerine fruit and reduces production efficiency. Summary of the invention

[0004] The purpose of the present invention is to provide a method and system for accurately predicting embryo sac parameters of fresh tangerine peel fruits, so as to facilitate the subsequent automatic decapsulation of fresh tangerine peel fruits.

[0005] To achieve the above-mentioned purpose, the present invention provides the following scheme: a method for predicting embryo sac parameters of tangerine peel fresh fruit, the method comprising: obtaining basic parameters of tangerine peel fresh fruit, measured parameters of embryo sac of tangerine peel fresh fruit and basic parameters of tangerine peel fresh fruit to be tested; the basic parameters of tangerine peel fresh fruit include: tangerine peel fresh fruit volume, tangerine peel fresh fruit weight, tangerine peel fresh fruit white volume and tangerine peel fresh fruit embryo sac volume; the measured parameters of embryo sac of tangerine peel fresh fruit include: embryo sac maximum radius, embryo sac height and embryo sac bottom distance; taking the corresponding basic parameters of multiple tangerine peel fresh fruits as input and the measured parameters of the multiple tangerine peel fresh fruits as output, training a neural network model; and using the trained neural network model to process the basic parameters of the tangerine peel fresh fruit to be tested to obtain corresponding embryo sac parameter prediction results.

[0006] Optionally, the method for obtaining the volume of the tangerine fresh fruit specifically includes: obtaining a dual-perspective image of the tangerine fresh fruit; preprocessing the dual-perspective image to obtain a preprocessed dual-perspective image; correcting a position offset error of the tangerine fresh fruit according to the preprocessed dual-perspective image to obtain a dual-perspective image after the position offset error is corrected; correcting the radius error of the dual-perspective image after the position offset error is corrected to obtain the actual radius of the tangerine fresh fruit; and calculating the volume of the tangerine fresh fruit using the actual radius of the tangerine fresh fruit.

[0007] Optionally, the specific method for obtaining the volume of the tangerine peel and the tangerine white of the tangerine peel and the volume of the embryo sac of the tangerine peel includes: obtaining the volume of the tangerine peel, the weight of the tangerine peel, the density of the tangerine peel and the density of the embryo sac of the tangerine peel; calculating the volume of the tangerine peel and the tangerine white of the tangerine peel and the volume of the embryo sac of the tangerine peel using the formula Vb and Vn; wherein ρ1 is the density of the tangerine peel and the tangerine white of the tangerine peel, ρ2 is the density of the embryo sac of the tangerine peel, Vall is the volume of the tangerine peel, and Mall is the weight of the tangerine peel.

[0008] Optionally, a weighing module with a cantilever beam structure is used to measure the weight of the fresh tangerine peel.

[0009] Optionally, a supervised learning method is used to train the neural network model.

[0010] A system for predicting embryo sac parameters of tangerine peel fresh fruit, the method comprising: a parameter acquisition unit, used for acquiring basic parameters of tangerine peel fresh fruit, measured parameters of embryo sac of tangerine peel fresh fruit and basic parameters of tangerine peel fresh fruit to be tested; the basic parameters of tangerine peel fresh fruit include: tangerine peel fresh fruit volume, tangerine peel fresh fruit weight, tangerine peel fresh fruit white volume and tangerine peel fresh fruit embryo sac volume; the measured parameters of embryo sac of tangerine peel fresh fruit include: embryo sac maximum radius, embryo sac height and embryo sac bottom distance; a prediction unit, used for training a neural network model with the corresponding basic parameters of a plurality of tangerine peel fresh fruits as input and the measured parameters of the plurality of tangerine peel fresh fruits as output; and using the trained neural network model to process the basic parameters of the tangerine peel fresh fruit to be tested, to obtain corresponding embryo sac parameter prediction results.

[0011] Optionally, the parameter acquisition unit includes a tangerine fresh fruit volume calculation subunit, which specifically includes: an image acquisition module, used to acquire a dual-perspective image of the tangerine fresh fruit; a preprocessing module, used to preprocess the dual-perspective image to obtain a preprocessed dual-perspective image; a position offset error correction module, used to correct the position offset error of the tangerine fresh fruit according to the preprocessed dual-perspective image to obtain a dual-perspective image after the position offset error is corrected; a radius error correction module, used to correct the radius error of the dual-perspective image after the position offset error is corrected to obtain the actual radius of the tangerine fresh fruit; and a tangerine fresh fruit volume calculation module, used to calculate the volume of the tangerine fresh fruit using the actual radius of the tangerine fresh fruit.

[0012] Optionally, the parameter acquisition unit also includes: a tangerine peel volume and tangerine white volume and tangerine peel embryo sac volume calculation subunit, which specifically includes: a parameter acquisition module, used to obtain the tangerine peel volume, the tangerine peel weight, the tangerine peel white density and the tangerine peel embryo sac density; a tangerine peel volume and tangerine peel embryo sac volume calculation module, used to calculate the tangerine peel volume Vb and the tangerine peel embryo sac volume Vn of the tangerine peel fresh fruit using a formula; wherein ρ1 is the density of the tangerine peel white of the tangerine peel fresh fruit, ρ2 is the density of the tangerine peel fresh fruit embryo sac, Vall is the tangerine peel volume, and Mall is the tangerine peel fresh fruit weight.

[0013] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects: the present invention calculates and solves the volume of the embryo sac and the volume of the tangerine peel fresh fruit according to the volume of the tangerine peel fresh fruit, the weight of the tangerine peel fresh fruit, the density of the tangerine peel and the density of the embryo sac of the tangerine peel fresh fruit, and obtains the embryo sac parameters of the tangerine peel fresh fruit by using a neural network model analysis. The prediction process will not destroy the shape of the tangerine peel fresh fruit, and the prediction result is accurate; according to the prediction result of the embryo sac parameters of the tangerine peel fresh fruit, the optimized processing tool feed parameters and tool expansion parameters can be generated to control the automated operation of rotary digging and removing the embryo sac, thereby improving the effect and speed of removing the sac. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0015] Figure 1 A schematic diagram of the process flow of the method for predicting embryo sac parameters of fresh tangerine peel fruit provided by the present invention; Figure 2 It is a longitudinal section image of the fresh orange fruit image taken by camera X; Figure 3 It is a rectangular image obtained from the longitudinal section image of the orange fresh fruit image taken by camera X; Figure 4 This is the principle diagram of distance position deviation correction; Figure 5 This is the principle diagram of radius error correction; Figure 6 Schematic diagram of the tangerine peel fresh fruit embryo sac parameter prediction system provided by the present invention.

[0016] Among them: 1-Camera X, 2-Camera Y, 3-Lens plane, 4-Image plane. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] The terms "first", "second", "third", etc. (if any) in the specification and claims of the present invention and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the objects described in this way can be interchanged under appropriate circumstances. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions.

[0019] In this patent document, the drawings discussed below and the embodiments used to describe the principles of the present invention are for illustration only and should not be interpreted as limiting the scope of the present invention. Those skilled in the art will understand that the principles of the present invention can be implemented in any appropriately arranged system. Exemplary embodiments will be described in detail, and examples of these embodiments are shown in the accompanying drawings. In addition, the terminal according to the exemplary embodiments will be described in detail with reference to the accompanying drawings. The same reference numerals in the accompanying drawings refer to the same elements.

[0020] The terms used in the present specification are only used to describe specific embodiments and are not intended to show the concept of the present invention. Unless there is a clear different meaning in the context, expressions used in the singular form include expressions in the plural form. In the present specification, it should be understood that terms such as "including", "having" and "containing" are intended to illustrate the possibility of the presence of features, numbers, steps, actions or combinations thereof disclosed in the present specification, and are not intended to exclude the possibility that one or more other features, numbers, steps, actions or combinations thereof may exist or may be added. The same reference numerals in the drawings refer to the same parts.

[0021] The purpose of the present invention is to provide a method and system for accurately predicting embryo sac parameters of fresh tangerine peel fruits, so as to improve the effect and speed of subsequent automatic decapsulation.

[0022] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0023] like Figure 1 As shown, it is a flow chart of a method for predicting embryo sac parameters of tangerine peel fresh fruit, the method comprising: step 101: obtaining basic parameters of tangerine peel fresh fruit, measured parameters of embryo sac of tangerine peel fresh fruit and basic parameters of tangerine peel fresh fruit to be tested; the basic parameters of tangerine peel fresh fruit include: volume of tangerine peel fresh fruit, weight of tangerine peel fresh fruit, volume of tangerine peel fresh fruit and volume of embryo sac of tangerine peel fresh fruit; the measured parameters of embryo sac of tangerine peel fresh fruit include: maximum radius of embryo sac, height of embryo sac and distance from bottom of embryo sac; step 102: taking the corresponding basic parameters of multiple tangerine peel fresh fruits as input and the measured parameters of the multiple tangerine peel fresh fruits as output, adopting a supervised learning method to train a neural network model; and using the trained neural network model to process the basic parameters of the tangerine peel fresh fruit to be tested, to obtain corresponding embryo sac parameter prediction results.

[0024] The present invention uses a neural network model to predict embryo sac parameters, thereby improving the accuracy and efficiency of the prediction.

[0025] For the weight of fresh tangerine peels, a cantilever beam structure based on a strain resistor bridge is used. When a heavy object is placed on the cantilever beam, it causes its deformation and then generates a corresponding change signal on the strain resistor bridge. After amplifying and processing the change signal data of the bridge, the weight of the fresh tangerine peels can be obtained, making the measured weight accurate.

[0026] Because the individual tangerine peel fresh fruit grows from an early spherical shape to an ellipsoid shape or a grapefruit shape with obvious head size, the difference is relatively large. In the process of predicting the embryo sac parameters, the influence of volume needs to be considered. Therefore, the measurement of the volume of tangerine peel fresh fruit is crucial. In order to improve the accuracy of volume measurement and expand the practicality of the volume measurement scheme, the present invention uses a dual-view machine vision method to measure the volume of tangerine peel fresh fruit. The measurement method specifically includes: obtaining a dual-view image of the tangerine peel fresh fruit; preprocessing the dual-view image to obtain a preprocessed dual-view image; correcting the position offset error of the tangerine peel fresh fruit according to the preprocessed dual-view image to obtain a dual-view image after the position offset error is corrected; correcting the radius error of the dual-view image after the position offset error is corrected to obtain the actual radius of the tangerine peel fresh fruit; and calculating the volume of the tangerine peel fresh fruit using the actual radius of the tangerine peel fresh fruit.

[0027] Cameras X and Y are located at the same height and 90 degrees outside the fresh fruit placement area. The corresponding distances to the center point of the tangerine peel fresh fruit placement are h1 and h2, which have been calibrated. Usually, tangerine peel fresh fruit is approximately spherical and ellipsoidal. For the ellipsoid with the sphere and the long axis perpendicular, the longitudinal section of the tangerine peel fresh fruit at the center point has the largest area in the corresponding direction.

[0028] The image of the tangerine fresh fruit after being photographed by camera X, without considering the measurement error, sphere radius error (sphere imaging radius occlusion error), misalignment error (distance error caused by the fresh fruit not being at the center point), fruit shape error (error caused by the fruit placement posture not being vertical and the fruit shape being asymmetric), the image data obtained by the camera is calculated through parameter calculation to obtain the image data of the longitudinal section of the tangerine fresh fruit as follows: Figure 2 , cut a small rectangle on the longitudinal section such as Figure 3 As shown, the parameters are the diameter and height of a cylinder. The longitudinal section is cut into multiple small rectangles, and the volume of the corresponding multiple small cylinders is added up to get the volume of the fresh tangerine fruit.

[0029] However, in actual applications, fresh tangerines may have an asymmetric shape, or be placed at a position other than the center point of the viewing angle, or be placed in a non-vertical position. A camera in a single position cannot correct the errors in the above situations. The above problems can also be solved by rotating the object to be measured to obtain a three-dimensional point cloud map of the target phenomenon. However, there are problems such as high cost, complex equipment, and time-consuming rotation. The use of two cameras with dual viewing angles can minimize the above errors.

[0030] Correction of asymmetric error: When the fruit shape is not a standard sphere or ellipsoid, one side is larger than the other side, the error will be large when the monocular camera takes one of the images for volume calculation. Using dual-viewing angles to intercept the corresponding radius data of the images from the two perspectives as the major and minor axes of the elliptical cylinder can reduce the asymmetric error of the fresh tangerine shape, and the required volume will be much more accurate.

[0031] Correction for distance position deviation: Figure 4 This is the principle diagram of distance position deviation correction. When the object position deviates from the set center point, the position of the image data (largest vertical profile) collected by it also deviates from the set center point, but the calculation process is still calculated according to the center point position, which will inevitably lead to calculation errors. When using a dual-view camera, the offset in the X direction can be evaluated based on the corresponding position of the image center point measured by camera Y, and the distance error of camera X can be corrected. Similarly, the offset in the Y direction can be evaluated based on the corresponding position of the image center point measured by camera X, and the distance error of camera Y can be corrected.

[0032] Correction of non-vertical errors: Since the monocular camera cannot collect depth information, when the fruit is not placed vertically but tilted, the image taken by a single camera cannot obtain the fruit's posture information. Since the fruit is tilted along the X-axis, the image collected on the X-axis cannot reflect the tilt of the fruit's posture, and the error with the real image is very large. The corresponding camera Y on the Y-axis collects the tilted posture of the fruit. By combining the images collected by the two cameras, the volume measurement error of the fruit in the tilted posture can be reduced.

[0033] Correction for radius error: When imaging a spherical object, since the light is irradiated along a straight line, it cannot pass through the point with the largest cross-section of the sphere. Usually, the projection of point A is taken as the maximum circumference point of the target, resulting in a measurement error. This error is called radius error. However, if the maximum interface distance h from the camera lens plane to the center point of the sphere is known, this error can be corrected.

[0034] like Figure 5 As shown, let the center point of the lens plane be O1 and the center point of the sphere be O, the distance between the two be h, the corresponding tangent point be A, the actual radius of the sphere be R, and the measured radius be r. After measuring and calculating the r value, R can be further solved to correct the radius error. In ΔAOO1, it is known that the length of OO1 is h and the length of AB is r, then it can be obtained: Two solutions will be generated when solving the equation. In practical applications, the distance h from the camera lens plane to the object is usually much larger than R. When the constraint condition h2>2R2 is satisfied, R can be obtained as: Calculating the volume of the tangerine peel fresh fruit according to the corrected tangerine peel fresh fruit radius can improve the accuracy of the volume calculation.

[0035] The specific method for obtaining the volume of the tangerine peel and the tangerine white of the tangerine peel and the volume of the embryo sac of the tangerine peel of the tangerine peel comprises: obtaining the volume of the tangerine peel, the weight of the tangerine peel, the density of the tangerine white of the tangerine peel and the density of the embryo sac of the tangerine peel; calculating the volume of the tangerine white of the tangerine peel and the volume of the embryo sac of the tangerine peel by using a formula; wherein ρ1 is the density of the tangerine white of the tangerine peel, ρ2 is the density of the embryo sac of the tangerine peel, Vall is the volume of the tangerine peel, and Mall is the weight of the tangerine peel.

[0036] The density of the tangerine white of the tangerine peel and the density of the embryo sac of the tangerine peel are important parameters of the tangerine peel. The volume of the tangerine white of the tangerine peel and the volume of the embryo sac of the tangerine peel calculated based on these important characteristic parameters are closer to the actual situation.

[0037] The invention corrects the volume of fresh tangerine peel fruit based on binocular machine vision, obtains weight parameters after amplifying weight signals using a weighing module with a cantilever beam structure, calculates the volume of tangerine peel fresh fruit and the volume of tangerine peel fresh fruit embryo sac according to density information, and processes the data using a neural network model to obtain accurate parameters of the embryo sac of the tangerine peel fresh fruit.

[0038] like Figure 6 As shown, the present invention also provides a system corresponding to the method for predicting embryo sac parameters of fresh tangerine peel fruit, the system comprising: a parameter acquisition unit 201 and a prediction unit 202.

[0039] The parameter acquisition unit 201 is used to acquire the basic parameters of tangerine peel fresh fruit, the measured parameters of tangerine peel fresh fruit embryo sac and the basic parameters of tangerine peel fresh fruit to be tested; the basic parameters of tangerine peel fresh fruit include: tangerine peel fresh fruit volume, tangerine peel fresh fruit weight, tangerine peel fresh fruit white volume and tangerine peel fresh fruit embryo sac volume; the measured parameters of tangerine peel fresh fruit embryo sac include: embryo sac maximum radius, embryo sac height and embryo sac bottom distance; the prediction unit 202 is used to train a neural network model with the corresponding basic parameters of multiple tangerine peel fresh fruits as input and the measured parameters of the multiple tangerine peel fresh fruits as output; and use the trained neural network model to process the basic parameters of the tangerine peel fresh fruit to be tested to obtain corresponding embryo sac parameter prediction results.

[0040] The parameter acquisition unit 201 includes: a tangerine peel fresh fruit volume calculation subunit and a tangerine peel fresh fruit tangerine white volume and tangerine peel fresh fruit embryo sac volume calculation subunit.

[0041] The tangerine peel fresh fruit volume calculation subunit specifically includes: an image acquisition module, used to acquire a dual-perspective image of the tangerine peel fresh fruit; a preprocessing module, used to preprocess the dual-perspective image to obtain a preprocessed dual-perspective image; a position offset error correction module, used to correct the position offset error of the tangerine peel fresh fruit according to the preprocessed dual-perspective image to obtain a dual-perspective image after the position offset error is corrected; a radius error correction module, used to correct the radius error of the dual-perspective image after the position offset error is corrected to obtain the actual radius of the tangerine peel fresh fruit; and a tangerine peel fresh fruit volume calculation module, used to calculate the volume of the tangerine peel fresh fruit using the actual radius of the tangerine peel fresh fruit.

[0042] The subunit for calculating the volume of tangerine peel and tangerine white and the volume of tangerine peel embryo sac specifically includes: a parameter acquisition module, used to obtain the volume of tangerine peel, the weight of tangerine peel, the density of tangerine peel and tangerine peel embryo sac density; a module for calculating the volume of tangerine peel and tangerine peel embryo sac volume, used to calculate the volume of tangerine peel and tangerine peel embryo sac volume Vb and the volume of tangerine peel embryo sac volume Vn using a formula; wherein ρ1 is the density of tangerine peel and tangerine white, ρ2 is the density of tangerine peel embryo sac, Vall is the volume of tangerine peel, and Mall is the weight of tangerine peel.

[0043] The prediction system provided by the present invention can also achieve the technical effect of the prediction method.

[0044] As for the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the description of the method part.

[0045] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A method for predicting embryo sac parameters of fresh tangerine peel fruit, characterized in that: The method comprises: Obtaining basic parameters of tangerine peel fresh fruit, measured parameters of tangerine peel fresh fruit embryo sac and basic parameters of tangerine peel fresh fruit to be measured; the basic parameters of tangerine peel fresh fruit include: tangerine peel fresh fruit volume, tangerine peel fresh fruit weight, tangerine peel fresh fruit white volume and tangerine peel fresh fruit embryo sac volume; the measured parameters of tangerine peel fresh fruit embryo sac include: embryo sac maximum radius, embryo sac height and embryo sac bottom distance; The method of obtaining the volume of the tangerine peel fresh fruit specifically comprises: taking the basic parameters corresponding to the plurality of tangerine peel fresh fruits as input and the measured parameters of the plurality of tangerine peel fresh fruits as output, training the neural network model; and processing the basic parameters of the tangerine peel fresh fruits to be tested by using the trained neural network model to obtain the corresponding embryo sac parameter prediction results. Acquire a dual-viewing angle image of the fresh tangerine fruit; Preprocessing the dual-view image to obtain a preprocessed dual-view image; According to the pre-processed dual-view image, the position offset error of the tangerine fresh fruit is corrected to obtain a dual-view image after the position offset error is corrected; Performing radius error correction on the dual-view image after the position offset error correction to obtain the actual radius of the tangerine fruit; The volume of the tangerine peel fresh fruit is calculated using the actual radius of the tangerine peel fresh fruit.

2. The method for predicting embryo sac parameters of fresh tangerine peel according to claim 1, characterized in that: The specific method for obtaining the volume of the tangerine peel fresh fruit and the volume of the tangerine peel fresh fruit embryo sac comprises: Obtaining the volume of the tangerine peel fresh fruit, the weight of the tangerine peel fresh fruit, the density of the tangerine peel white, and the density of the tangerine peel fresh fruit embryo sac; Using the formula Calculate the volume V of the orange red fresh fruit and orange white b and the tangerine peel fresh fruit embryo sac volume V n ; Among them, ρ1 is the density of the orange peel of fresh orange fruit, ρ2 is the density of the embryo sac of fresh orange fruit, V all is the volume of fresh tangerine fruit, M all It is the weight of fresh tangerine fruit.

3. The method for predicting embryo sac parameters of tangerine peel fresh fruit according to claim 1, characterized in that: The weight of the fresh tangerine peel is measured by utilizing a weighing module with a cantilever beam structure.

4. The method for predicting embryo sac parameters of tangerine peel fresh fruit according to claim 1, characterized in that: The neural network model is trained using a supervised learning method.

5. A system for predicting embryo sac parameters of tangerine peel fresh fruit according to the method for predicting embryo sac parameters of tangerine peel fresh fruit according to any one of claims 1 to 4, characterized in that: The system comprises: The parameter acquisition unit is used to acquire the basic parameters of the tangerine peel fresh fruit, the measured parameters of the tangerine peel fresh fruit embryo sac and the basic parameters of the tangerine peel fresh fruit to be tested; the basic parameters of the tangerine peel fresh fruit include: the volume of the tangerine peel fresh fruit, the weight of the tangerine peel fresh fruit, the volume of the tangerine peel fresh fruit and the volume of the tangerine peel fresh fruit embryo sac; the measured parameters of the tangerine peel fresh fruit embryo sac include: the maximum radius of the embryo sac, the height of the embryo sac and the distance from the bottom of the embryo sac; The prediction unit is used to train the neural network model with the basic parameters corresponding to the plurality of tangerine peel fresh fruits as input and the measured parameters of the plurality of tangerine peel fresh fruits as output; and use the trained neural network model to process the basic parameters of the tangerine peel fresh fruits to obtain the corresponding embryo sac parameter prediction results.

6. A tangerine peel fresh fruit embryo sac parameter prediction system according to claim 5, characterized in that: The parameter acquisition unit includes a tangerine fresh fruit volume calculation subunit, which specifically includes: An image acquisition module, used for acquiring dual-viewing angle images of the tangerine fresh fruit; A preprocessing module, used for preprocessing the dual-view image to obtain a preprocessed dual-view image; A position offset error correction module, used to correct the position offset error of the tangerine fresh fruit according to the pre-processed dual-view image, to obtain a dual-view image after the position offset error is corrected; A radius error correction module, used to perform radius error correction on the dual-view image after the position offset error correction, to obtain the actual radius of the tangerine fruit; The tangerine peel fresh fruit volume calculation module is used to calculate the tangerine peel fresh fruit volume using the actual radius of the tangerine peel fresh fruit.

7. A tangerine peel fresh fruit embryo sac parameter prediction system according to claim 5 or 6, characterized in that: The parameter acquisition unit further includes: a subunit for calculating the volume of tangerine peel fresh fruit and tangerine white and tangerine peel fresh fruit embryo sac volume, and the subunit specifically includes: A parameter acquisition module, used for acquiring the volume of the tangerine peel fresh fruit, the weight of the tangerine peel fresh fruit, the density of the tangerine peel white and the density of the tangerine peel fresh fruit embryo sac; The module for calculating the volume of fresh orange white and embryo sac of orange red fruit is used to calculate the volume of fresh orange white and embryo sac of orange red fruit using the formula Calculate the volume V of the orange red fresh fruit and orange white b and the tangerine peel fresh fruit embryo sac volume V n ; Among them, ρ1 is the density of the orange peel of fresh orange fruit, ρ2 is the density of the embryo sac of fresh orange fruit, V all is the volume of fresh tangerine fruit, M all It is the weight of fresh tangerine fruit.

Images