Method and device for controlling a fruit picking robot, electronic device

Abstract

The application relates to a control method and device of a fruit sorting robot and electronic equipment. The method is applied to a fruit sorting system, the fruit sorting system comprises a fruit sorting robot and a conveying belt, the fruit sorting robot comprises a visual sensor, an infrared sensor and a mechanical arm, the method comprises the following steps: collecting an image of fruits to be sorted on the conveying belt by using the visual sensor; analyzing the image to determine the position and type of each fruit to be sorted; calling the mechanical arm to group fruits of the same type in the same path of the conveying belt based on the position, conveying speed of the conveying belt and type; controlling the near-infrared sensor to emit near-infrared rays to the fruits to be sorted in each path according to the type; and determining the quality grade of the fruits to be sorted according to the type, the emitted near-infrared rays and the analyzed received near-infrared rays for each path.

Description

Technical Field

[0001] This application relates to the field of robot control technology, and in particular to a control method, device, and electronic equipment for a fruit sorting robot. Background Technology

[0002] Fruits have great benefits for the human body, and people's demand for fruits is increasing. As a result, more and more fruit markets are being established. Fruit markets receive large quantities of fruits at once, and then the fruits are sorted by hand. People use their experience to select fruits based on quality, and choose the fruits with the best appearance and quality for sale.

[0003] However, with the expansion of the fruit market and the increasing quantity of fruit received each time, the number of personnel required for quality inspection also increases. Furthermore, manual quality inspection of fruit is prone to significant errors, and inconsistent standards and subjective biases among inspectors lead to inaccurate results. Therefore, currently, the common practice is to manually inspect fruit for damage and classify them by type, then randomly select a portion of the fruit as samples. A portion of the pulp from these samples is then cut and analyzed using machines to determine the composition of the pulp and thus the quality of the entire batch. However, this method damages the sample fruit and cannot accurately determine the quality of each individual fruit. Summary of the Invention

[0004] This application provides a control method, device, and electronic equipment for a fruit sorting robot, aiming to solve the current problem of accurately and quickly identifying the quality of fruit without damaging it.

[0005] In a first aspect, this application provides a control method for a fruit sorting robot, applied to a fruit sorting system, the fruit sorting system comprising: a fruit sorting robot and a conveyor belt, the fruit sorting robot comprising: a vision sensor, a near-infrared sensor, and a robotic arm, the method comprising:

[0006] Images of fruits to be sorted on a conveyor belt are captured using a vision sensor.

[0007] Analyze the image to determine the location and type of each fruit in the fruit to be sorted;

[0008] Based on the location, the conveyor belt speed, and the type, a robotic arm is invoked to group fruits of the same type into the same path on the conveyor belt.

[0009] According to the type, the near-infrared sensor is controlled to emit near-infrared rays toward the fruit to be sorted in each path;

[0010] For each path, the quality grade of the fruit to be sorted is determined based on the type, the emitted near-infrared light, and the received near-infrared light.

[0011] The scheme obtains the fruit type and specific position of the fruit sent to the conveying belt by analyzing the image shot by the visual sensor, and then the same type of fruit is sorted into the same path conveying belt by the mechanical arm, so that the fruit is automatically classified and divided, avoiding the manual sorting process. Then, the near-infrared sensor emits near-infrared rays to the fruit, and further analyzes the received near-infrared rays to obtain the internal composition of each fruit. Finally, the fruit is classified according to the pre-set rules, realizing the use of robots to classify and select the quality of a large number of fruits without damage, saving labor cost and being more accurate in fruit classification.

[0012] Optionally, the near-infrared sensor comprises: a plurality of light source emitters and a plurality of light source receivers.

[0013] The method comprises the following steps:

[0014] According to the type, the average size, the flesh thickness, and the sugar content of the type are obtained.

[0015] According to the size, a corresponding number of light source emitters and light source receivers are started.

[0016] According to the flesh thickness and the sugar content, the detection mode of the near-infrared sensor is determined.

[0017] The scheme determines the average size, the flesh thickness, the sugar content, and other properties of the type of fruit to be sorted, and then starts a corresponding number of near-infrared sensors according to the size, and adjusts the detection mode of the near-infrared sensor to the fruit to be sorted according to the flesh thickness and the sugar content, so that the most appropriate detection mode can be adopted for the fruit to be sorted at all times, and the most accurate composition data of the fruit to be sorted can be obtained.

[0018] Optionally, according to the type, the emitted near-infrared rays, and the received near-infrared rays, the quality grade of the fruit to be sorted is determined, comprising:

[0019] According to the type, the pre-set analysis spectrum corresponding to the type is determined.

[0020] According to the emitted near-infrared rays and the received near-infrared rays, the near-infrared absorption data of the fruit to be sorted is determined.

[0021] The pre-set analysis spectrum and the near-infrared absorption data are analyzed to determine the ingredient content of the fruit to be sorted.

[0022] According to the pre-set standard and the ingredient content, the quality grade of the fruit to be sorted is determined.

[0023] By the scheme, there is a corresponding preset analysis spectrum for each fruit type, so that each fruit has an accurate comparison, and then the near-infrared radiation of the fruit to be sorted is emitted and received, the absorption data of each component of the fruit to the near-infrared is determined, the absorption data of the fruit to be sorted to different wavelength near-infrared is corresponded to the internal component of the fruit to be sorted, so that the internal component of the fruit to be sorted is obtained in a non-destructive manner, and then the quality of the fruit to be sorted is determined according to the content of the internal component.

[0024] Optionally, the near-infrared absorption data of the fruit to be sorted is determined according to the emitted near-infrared and the received near-infrared, and includes:

[0025] The light intensity information of the emitted near-infrared is obtained according to the emitted near-infrared;

[0026] The light intensity information of the received near-infrared is obtained according to the received near-infrared;

[0027] The near-infrared absorption data of the near-infrared is determined according to the light intensity information of the emitted near-infrared and the light intensity information of the received near-infrared.

[0028] By the scheme, according to the characteristics that different components of the fruit to be sorted have different absorption data to different wavelength near-infrared, different wavelength near-infrared is emitted to the fruit to be sorted, and then the change of the light intensity information of the near-infrared passing through the fruit to be sorted is analyzed to determine the near-infrared absorption data of the near-infrared.

[0029] Optionally, the position and type of each fruit in the fruit to be sorted are determined according to the image, and include:

[0030] Target recognition is performed on the image, and the conveying belt in the image is regionally divided;

[0031] The position of each region relative to the conveying belt is determined;

[0032] The fruit type in each region is obtained according to the image.

[0033] By the scheme, the picture target recognition is performed to regionally divide the conveying belt, and the position of each region relative to the conveying belt is analyzed and determined, and then the fruit type in each region is analyzed one by one, the position of the region is represented as the position of the fruit, and the speed of obtaining the position of the fruit is improved.

[0034] Optionally, based on the position, the conveying speed of the conveying belt, and the type, a mechanical arm is called to group fruits of the same type in the same path of the conveying belt, and includes:

[0035] According to the position, the grabbing position of the mechanical arm is adjusted to the position;

[0036] According to the conveying speed of the conveying belt, the image shooting time, the time when the position is determined to be conveyed to the mechanical arm grabbing position;

[0037] According to the time, the mechanical arm is controlled to grab the fruit at the position;

[0038] According to the category, the fruit is placed into a designated path.

[0039] The scheme can obtain the time when the conveying belt is conveyed to the position according to the conveying speed of the conveying belt and the image shooting time, so as to adjust the grabbing time of the mechanical arm, and the mechanical arm can be adjusted to be positioned in advance at the position where the fruit to be sorted will be conveyed in the future according to the position of each fruit, so that the mechanical arm can accurately grab each fruit to be sorted, and the fruits are placed into the same path according to the category of the fruit, so as to complete the classification of the fruits to be sorted.

[0040] Optionally, the category of the fruit in each region is obtained according to the image, including:

[0041] The shape, size, and peel color of the fruit in each frame of picture are compared with the database features, and the category of the fruit is matched;

[0042] If the matching result is multiple results, the visual sensor is called to shoot a picture of the seeds inside the fruit;

[0043] The picture features of the seeds are analyzed, compared with seed example pictures corresponding to the multiple results, and the category of the fruit is determined.

[0044] The picture analysis technology is used to analyze the shape, size, peel color, and other features of the fruit to be sorted, and then the database is searched according to the features to preliminarily obtain the category of the fruit to be sorted. The category of the fruit to be sorted can be accurately obtained by combining the features of the fruit to be sorted, and when the category cannot be determined as one, the picture of the seeds of the fruit to be sorted is compared with the seed example picture to accurately obtain the category of the fruit to be sorted, thereby improving the accuracy of obtaining the fruit to be sorted.

[0045] In a second aspect, the application provides a control device of a fruit sorting robot, which is applied to a fruit sorting system, the fruit sorting system including a fruit sorting robot and a conveying belt, the fruit sorting robot including a visual sensor, a near-infrared sensor, and a mechanical arm, and the device including:

[0046] The acquisition module is configured to acquire, by using the visual sensor, an image of a fruit to be sorted on the conveying belt;

[0047] The analysis module is configured to analyze the image to determine the position and category of each fruit to be sorted.

[0048] a categorizing module configured to invoke a robot arm to categorize fruits of the same type into the same path of the conveyor belt based on the position, the conveying speed of the conveyor belt, and the type;

[0049] a detecting module configured to control the near-infrared sensor to emit near-infrared rays to the fruits to be sorted in each path according to the type;

[0050] a quality module configured to determine a quality grade of the fruits to be sorted according to the type, the emitted near-infrared rays, and the received near-infrared rays for each path.

[0051] Optionally, the near-infrared sensor comprises a plurality of light source emitters and a plurality of light source receivers; when the detecting module controls the near-infrared sensor to emit near-infrared rays to the fruits to be sorted in each path according to the type, the detecting module is specifically configured to:

[0052] obtain an average size, a flesh thickness, and a sugar content of the type according to the type;

[0053] start a corresponding number of the light source emitters and the light source receivers according to the size;

[0054] determine a detection mode of the near-infrared sensor according to the flesh thickness and the sugar content.

[0055] Optionally, when the quality module determines the quality grade of the fruits to be sorted according to the type, the emitted near-infrared rays, and the received near-infrared rays, the quality module is specifically configured to:

[0056] determine a preset analysis spectrum corresponding to the type according to the type;

[0057] determine near-infrared absorption data of the fruits to be sorted according to the emitted near-infrared rays and the received near-infrared rays;

[0058] analyze the preset analysis spectrum and the near-infrared absorption data to determine a component content of the fruits to be sorted;

[0059] determine the quality grade of the fruits to be sorted according to a preset standard and the component content.

[0060] Optionally, when the quality module determines the near-infrared absorption data of the fruits to be sorted according to the emitted near-infrared rays and the received near-infrared rays, the quality module is specifically configured to:

[0061] obtain emitted light intensity information according to the emitted near-infrared rays;

[0062] obtain received light intensity information according to the received near-infrared rays;

[0063] determining near-infrared absorption data of the near-infrared rays according to the emitted light intensity information and the received light intensity information.

[0064] Optionally, when the analysis module determines the position and the type of each fruit in the fruits to be sorted according to the image, the analysis module is specifically configured to:

[0065] performing target recognition on the image, and dividing the conveying belt in the image into regions;

[0066] determining the position of each region relative to the conveying belt;

[0067] acquiring the type of the fruit in each region according to the image.

[0068] Optionally, when the classification module classifies the fruits of the same type into the same path of the conveying belt based on the position, the conveying speed of the conveying belt, and the type by calling the mechanical arm, the classification module is specifically configured to:

[0069] adjusting the grabbing position of the mechanical arm to the position according to the position;

[0070] determining the time when the position is conveyed to the grabbing position of the mechanical arm according to the conveying speed of the conveying belt and the shooting time of the image; and

[0071] controlling the mechanical arm to grab the fruit at the position according to the time.

[0072] placing the fruit into a designated path according to the type.

[0073] Optionally, when the analysis module acquires the type of the fruit in each region according to the image, the analysis module is specifically configured to:

[0074] comparing the shape, size, and peel color of the fruit in each frame of picture with the features in a database, and matching the type of the fruit;

[0075] if the matching result is multiple results, calling the visual sensor to shoot a picture of the seeds inside the fruit;

[0076] analyzing the picture features of the seeds, and comparing the picture features with seed example pictures corresponding to the multiple results, to determine the type of the fruit.

[0077] In a third aspect, the present application provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program capable of being loaded and executed by the processor to execute the method of the first aspect.

[0078] In a fourth aspect, the present application provides a computer readable storage medium, which stores a computer program capable of being loaded and executed by a processor to execute the method of the first aspect.

[0079] In a fifth aspect, the present application provides a fruit sorting system, comprising: a fruit sorting robot, a conveyor belt, the fruit sorting robot comprising: a visual sensor, a near-infrared sensor, a mechanical arm;

[0080] the conveyor belt, configured to convey fruits to be sorted; the fruit sorting robot, configured to: collect an image of the fruits to be sorted on the conveyor belt by using the visual sensor; analyze the image to determine a position and a kind of each of the fruits to be sorted; based on the position, a conveying speed of the conveyor belt, and the kind, call the mechanical arm to group fruits of the same kind in a same path of the conveyor belt; control the near-infrared sensor to emit near-infrared rays to the fruits to be sorted in each path according to the kind; and determine a quality grade of the fruits to be sorted in each path according to the kind, the emitted near-infrared rays, and received near-infrared rays. BRIEF DESCRIPTION OF DRAWINGS

[0081] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or prior art description will be briefly introduced. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0082] Figure 1 An application scenario schematic diagram provided by an embodiment of the present application;

[0083] Figure 2 A flowchart of a control method of a fruit sorting robot provided by an embodiment of the present application;

[0084] Figure 3 A cross-sectional view of a near-infrared sensor provided by an embodiment of the present application;

[0085] Figure 4 A structural schematic diagram of a control device of a fruit sorting robot provided by an embodiment of the present application;

[0086] Figure 5 A structural schematic diagram of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0087] In order to make the purposes, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0088] In addition, the term "and / or" in this document merely describes an association relationship of associated objects, and indicates that there can be three relationships, for example, A and / or B can represent three cases of A existing alone, A and B existing simultaneously, and B existing alone. In addition, the character " / " in this document generally represents an "or" relationship between the front and rear associated objects unless otherwise specified.

[0089] The embodiments of the present application will be described in further detail below with reference to the drawings of the specification.

[0090] With the expansion of the fruit market, the number of fruits purchased by large supermarkets or markets at a time is increasing, and how to strictly screen the quality of fruits has become a more and more serious problem. At present, the quality of fruits is mainly detected from several aspects such as damage of fruit appearance, sugar content of fruits, and the rating of fruits is also mainly referred to these factors. It is common to distinguish whether the appearance of fruits is damaged by artificial visual inspection, to classify fruits according to types, and then to select part of each type of fruits as samples, to cut part of the flesh of the sample fruits, and to use a machine to detect the composition of the sample flesh to determine the quality of a batch of fruits. However, this detection method will damage the sample fruits, and cannot accurately know the quality of each fruit.

[0091] Based on this, the present application provides a control method and device of a fruit sorting robot, and an electronic equipment, which are applied to the fruit sorting robot, so that the fruit sorting robot can realize nondestructive quality screening of fruits.

[0092] Figure 1 An application scenario provided by the present application is shown in FIG. 1, which is a top view of the sorting process of the entire sorting system. A rectangle 10 is a conveying belt, a black square 11 is a fruit to be sorted, a white square 12 is a fruit sorting robot, and a dotted square 13 is a position where the fruit to be sorted will be placed. Figure 1 During the transportation of the fruit to be sorted by the conveying belt, the sorting robot grasps the fruit to be sorted, classifies and monitors the quality of the fruit to be sorted, and then places the fruit to be sorted on another conveying belt, thereby completing the sorting and quality monitoring of the fruit to be sorted.

[0093] Figure 2A flow chart of a control method of a fruit sorting robot is provided for an embodiment of the present application. The method of the embodiment can be applied to a chip of a sorting robot in the above scenario. The sorting system includes a conveyor belt conveying fruits and a fruit sorting robot. The fruit sorting robot includes a vision sensor, an infrared sensor, a mechanical arm, and a chip integrating the method. The chip is installed inside the robot, and the vision sensor, the infrared sensor, and the mechanical arm are fixedly installed on the robot. As shown in FIG. 8, the method includes the following steps. Figure 2

[0094] S201, collecting an image of a fruit to be sorted on the conveyor belt by using a vision sensor.

[0095] The vision sensor is a device for obtaining an image, including but not limited to a camera.

[0096] The fruit to be sorted is all fruits that need to be sorted and determined for quality. The fruits to be sorted can be multiple types of fruits sorted and determined for quality together, or a single type of fruit sorted and determined for quality.

[0097] In some implementations, the vision sensor is called to take an image of the fruit to be sorted conveyed on the conveyor belt. The shooting mode can be to take one picture at a fixed time, and the fixed time is the time for the part of the conveyor belt to be photographed to move out of the shooting area.

[0098] S202, analyzing the image to determine the position and type of each fruit in the fruit to be sorted.

[0099] The position indicates the position of the fruit relative to the conveyor belt, such as a position 10 cm from the left edge and 10 cm from the top edge of the conveyor belt in the picture. In some implementations, if the image collection range of the vision sensor covers the entire conveyor belt, the edge of the conveyor belt in the picture is the boundary of the conveyor belt; in other implementations, if the image collection range of the vision sensor covers part of the conveyor belt, the edge of the conveyor belt in the picture is the part of the picture that belongs to the conveyor belt.

[0100] In some implementations, the image recognition technology is used to recognize the image, select all the fruits in the image, and determine the position of each fruit. Specifically, the distance between each fruit in the picture and the edge of the conveyor belt, and the ratio of the picture conveyor belt to the real conveyor belt can be used to calculate the distance between the real fruit and the edge of the conveyor belt.

[0101] ​The kind of fruit can also be identified based on the trained image recognition model. Specifically, a large number of fruit pictures can be used as samples, and the corresponding fruit kind can be used as a label to train the image recognition model. The trained model can accurately extract the features of the fruit in the image during image recognition, and compare the features with the features of various kinds of fruit to determine the kind of fruit in the image. For example, if the fruit is fresh red, the maximum diameter is 3 cm, the achenes are sharp and oval, and there are many black spots, it can be determined that the fruit is a strawberry.

[0102] S203, based on the position, the conveying speed of the conveying belt, and the kind, calling the mechanical arm to group the fruits of the same kind in the same path of the conveying belt.

[0103] In one implementation, the structure of the conveying belt can include one main conveying belt and multiple sub-conveying belts, each sub-conveying belt being a path, and the main conveying belt and the sub-conveying belts being adjacent in the conveying direction and seamlessly spliced together. During operation, the fruits can be conveyed from the main conveying belt to the sub-conveying belts.

[0104] In another implementation, the structure of the conveying belt is one conveying belt and multiple baffles on the conveying belt, and the conveying belt is divided into multiple regions of the same length as the conveying belt by the baffles, each region being a path.

[0105] Specifically, for the specific position of each fruit to be sorted relative to the conveying belt, the mechanical arm can be controlled to grasp the fruit to be sorted at that position, and the moving speed of the mechanical arm can be adjusted according to the speed of the conveying belt, so that the mechanical arm can accurately grasp the fruit during the movement of the conveying belt, and then place the fruit on the corresponding path.

[0106] S204, according to the kind, controlling the near-infrared sensor to emit near-infrared rays to the fruit to be sorted in each path.

[0107] The near-infrared sensor includes a light source emitter and a light source receiver. The structure of the near-infrared sensor is shown in Figure 3 Figure 3 is a cross-sectional view of the near-infrared sensor, and the infrared sensor has three light source emitters 31 and three light source receivers 32.

[0108] Specifically, according to the kind of fruit to be sorted, the specific way of sending near-infrared rays by the near-infrared sensor is determined, and then the light source emitter of the near-infrared sensor is controlled to emit infrared rays of different wavelengths to each fruit in the path.

[0109] S205, for each path, according to the kind, the emitted near-infrared rays, and the received near-infrared rays, determining the quality grade of the fruit to be sorted.

[0110] ​The quality grade refers to the classification of fruits of the same variety according to the sugar content, acidity, internal lesions and other factors.

[0111] Specifically, based on the received near-infrared light, light intensity information of the received near-infrared light is obtained, and then based on the light intensity information of the near-infrared light emitted by the near-infrared sensor, light intensity change information of the near-infrared light is obtained.

[0112] Then, according to the type of the fruit to be sorted, the light intensity absorption standard of the fruit to be sorted is determined, and then the light intensity absorption standard is compared with the light intensity change information, so that the sugar content, acidity and lesion condition of the fruit to be sorted can be determined. Then, based on the sugar content, acidity and lesion condition, the quality grade of the fruit to be sorted is classified.

[0113] The fruit type and specific position of the fruit sent to the conveying belt are obtained by analyzing the image captured by the visual sensor, and then the same type of fruit is sorted into the same path of the conveying belt by the mechanical arm, so that the fruit is automatically classified and divided, avoiding the manual sorting process. Then, the near-infrared sensor emits near-infrared light to the fruit, and further analyzes the received near-infrared light to obtain the internal composition of each fruit. Finally, the fruit is classified according to the pre-set rules, realizing the use of robots to classify and select the quality of a large number of fruits without damage, saving labor costs and improving the accuracy of fruit classification.

[0114] In some embodiments, the sorting system further comprises a marking machine, and after the step S205, the quality information of the fruit to be sorted is sent to the marking machine, the marking machine generates a two-dimensional code of the quality information, and prints the two-dimensional code on a label, and then the label is attached to the fruit to be sorted.

[0115] In some embodiments, the near-infrared sensor in the fruit sorting system is located in each path of the conveying belt, and one near-infrared sensor is responsible for the detection of fruits in one path. The sorting robot and the near-infrared sensor are in communication connection.

[0116] In some other embodiments, the sorting system further comprises an automatic boxing machine, and the automatic boxing machine is located at the end of the conveying belt. When the fruit to be sorted is conveyed to the end of the conveying belt, it will automatically fall off. The automatic boxing machine contains a plurality of grooves, so that when the fruit falls off, it falls into the groove of the automatic boxing machine. After each groove is filled with fruit, the position of the automatic boxing machine is adjusted according to the speed of the conveying belt and the time when the fruit sorting robot places the fruit to be sorted in the path, and the position of the fruit to be sorted, so that the new groove is located at the position of the next fruit falling off. In some implementations, each path of the conveying belt is further divided according to the quality grade. When the automatic boxing machine automatically boxes the fruit, one automatic boxing machine boxes each fruit according to the quality grade.

[0117] In some embodiments, when the fruit to be sorted is only one kind of fruit, the visual sensor does not detect the kind of fruit in step S202, but only uses image recognition to determine whether the outer surface of the fruit to be sorted is damaged, and if the outer surface of the fruit to be sorted is damaged, it is directly determined as low quality, and each path corresponds to each quality level of the fruit to be sorted.

[0118] In some embodiments, the near-infrared sensor includes a plurality of light source emitters and a plurality of light source receivers. The control of the near-infrared sensor to emit near-infrared rays to the fruit to be sorted in each path according to the kind includes: obtaining the average size, flesh thickness, and sugar content of the kind; starting a corresponding number of light source emitters and light source receivers according to the size; and determining the detection mode of the near-infrared sensor according to the flesh thickness and the sugar content.

[0119] The detection mode of the near-infrared sensor includes reflection detection and transmission detection.

[0120] It should be noted that the near-infrared sensor is composed of a plurality of light source emitters and light source receivers. In transmission detection, the positions of the light source emitters and the light source receivers are adjusted so that the light source emitters and the light source receivers are close to the peel of the fruit to be sorted. In reflection detection, the light source emitters are at a certain distance from the fruit to be sorted.

[0121] Specifically, after obtaining the kind of fruit to be sorted, the corresponding size, flesh thickness, and sugar content of each kind of fruit are obtained from the fruit database.

[0122] Then, it is determined which gradient of the preset standard the size is located in to determine the number of light source emitters to be started. For example, a watermelon has an average diameter of 27 cm, and three light sources need to be started. An apple has an average diameter of 8 cm, and two light sources need to be started. The preset standard is a gradient size standard, and each gradient corresponds to a number of light source emitters. The size is located in a corresponding interval, and a corresponding number of light source emitters are started. According to the number, it is determined which position of the light source emitters to start, such as one light source emitter above the fruit to be sorted, two light source emitters on the left and right of the fruit to be sorted.

[0123] For the flesh thickness and the sugar content, the detection mode of the near-infrared sensor is determined. When the flesh thickness is too thick, the near-infrared rays cannot pass through the entire fruit body of the fruit to be sorted in reflection detection, so that the composition of the entire fruit flesh inside the fruit to be sorted cannot be obtained. Therefore, transmission detection is started so that the near-infrared rays can pass through the entire fruit body of the fruit to be sorted. When the sugar content is too low, the detection accuracy may be reduced in reflection detection, and transmission detection is also needed.

[0124] When either the flesh thickness or the sugar content reaches the need to start the transmission detection, directly start the transmission detection.

[0125] The scheme determines the average size, flesh thickness, sugar content and other attributes of the kind of fruit to be sorted, and then enables a corresponding number of near-infrared sensors according to the size, and adjusts the detection method of the near-infrared sensor for the fruit to be sorted according to the flesh thickness and sugar content, so that the most appropriate detection method can be adopted for the fruit to be sorted at all times, and the most accurate component data of the fruit to be sorted is obtained.

[0126] In some embodiments, according to the kind, emitted near-infrared rays, and analyzed received near-infrared rays, the quality grade of the fruit to be sorted is determined, including: according to the kind, determining a preset analysis spectrum corresponding to the kind; according to the emitted near-infrared rays and the received near-infrared rays, determining near-infrared absorption data of the fruit to be sorted; analyzing the preset analysis spectrum and the near-infrared absorption data to determine the component content of the fruit to be sorted; and according to a preset standard and the component content, determining the quality grade of the fruit to be sorted.

[0127] The establishment process of the preset analysis spectrum is that the near-infrared rays are used to measure each quality of each kind of fruit, in each measurement, the fruit is measured at different positions and the average value is taken to obtain the quality spectrum information of each kind of fruit as the preset analysis spectrum, and then the kind of fruit to be sorted and the preset analysis spectrum are stored in the analysis model library.

[0128] The preset standard is a standard set for the content of each component in each kind of fruit. For example, the water content and the sugar content.

[0129] After the kind of fruit to be sorted is determined, the preset analysis spectrum corresponding to the kind is selected by screening in the analysis model library.

[0130] According to the receiving method of the near-infrared rays, the analysis method of the near-infrared rays is determined, if the near-infrared rays are received by transmission, the transmission analysis method of the near-infrared rays is adopted to generate a transmission infrared spectrum, and if the near-infrared rays are received by reflection, the reflection analysis method of the near-infrared rays is adopted to generate a reflection infrared spectrum.

[0131] The preset analysis spectrum and the near-infrared spectrum are compared to determine the information of absorption of the near-infrared light by the fruit to be sorted under different wavelengths of the near-infrared light. According to the component type of the fruit to be sorted corresponding to each wavelength, the absorption information of the fruit to be sorted to each wavelength is analyzed, and then the absorption information is compared with the preset analysis spectrum to determine the content of each component in the fruit to be sorted. Then, the content of the component is compared with the content standard of the component in the preset standard to judge the quality grade of the fruit to be sorted. For example, the absorption information of the water honey peach at 1450 nm corresponds to water. According to the absorption of the water honey peach at 1450 nm, the proportion of the internal water of the water honey peach can be determined, and then the quality of the water honey peach is determined. For example, if the absorption rate is less than 15%, it can be determined that the water honey peach has less water and low quality.

[0132] According to the scheme, there is a corresponding preset analysis spectrum for each fruit type, so that each fruit has accurate comparison. Then, the fruit to be sorted is radiated and received by the near-infrared light to determine the near-infrared absorption data of each component of the fruit, and the absorption data of the fruit to be sorted to different wavelengths of the near-infrared light corresponds to the internal component of the fruit to be sorted, so that the internal component of the fruit to be sorted is obtained in a non-destructive manner, and then the quality of the fruit to be sorted is determined according to the content of the internal component.

[0133] In some embodiments, the quality grade of the fruit to be sorted is determined according to the type, the emitted near-infrared light, and the received near-infrared light, including: obtaining the emitted light intensity information according to the emitted near-infrared light; obtaining the received light intensity information according to the received near-infrared light; and determining the near-infrared absorption data of the near-infrared light according to the emitted light intensity information and the received light intensity information.

[0134] The light intensity information includes wavelength information, frequency information, and optical density information, and the near-infrared absorption data is the information of the loss of optical density of the near-infrared light after passing through the fruit.

[0135] Since the reactions of different components in the fruit to different wavelengths of the near-infrared light are obviously different, the near-infrared sensor continuously emits near-infrared light of different wavelengths to irradiate the fruit to be sorted, the fruit to be sorted selectively absorbs the near-infrared light of different wavelengths, and the near-infrared light is weakened in some wavelength ranges after absorption, and then the near-infrared sensor receives the near-infrared light passing through the fruit to be sorted.

[0136] For the same wavelength of the near-infrared light, the optical density of the emitted near-infrared light when emitted is compared with the optical density of the near-infrared light when received to determine the near-infrared absorption data of the fruit to be sorted to each wavelength of the near-infrared light.

[0137] By the scheme, different wavelengths of near-infrared rays are emitted to the fruit to be sorted according to the characteristics that different components of the fruit to be sorted have different absorption data of different wavelengths of near-infrared rays, and then the change of the light intensity information of the near-infrared rays passing through the fruit to be sorted is analyzed to determine the near-infrared absorption data of the near-infrared rays.

[0138] In some embodiments, determining the position and the kind of each fruit in the fruit to be sorted according to the image comprises: performing target recognition on the image, and dividing the conveying belt in the image into regions; determining the position of each region relative to the conveying belt; and obtaining the kind of the fruit in each region according to the image.

[0139] The position represents the position of the fruit to be sorted relative to the conveying belt.

[0140] In some implementations, the image recognition technology is used to distinguish the conveying belt from other environments in the scene, and then the length of each edge of the conveying belt and the relative distance between the edges are collected. Then, the conveying belt in the picture is divided into square regions with equal side lengths, and each region is assigned a unique number.

[0141] For each region, the distance of the region relative to the four edges of the conveying belt is calculated. The distance of each region relative to the edges of the conveying belt is calculated as follows: the distance of the left edge of the region relative to the left edge of the conveying belt, and the distances of the four edges of the region relative to the four edges of the conveying belt are calculated in the same way. The distance of each region relative to the edges of the conveying belt is the position of each region.

[0142] Then, the number of the region covered by each fruit to be sorted is determined, and the number of the region where the edge of the fruit to be sorted is located is further determined. The position of the region where the most edge of the fruit to be sorted is located is taken as the position of the fruit to be sorted. For example, an apple covers 50 regions, and the regions where the leftmost, uppermost, rightmost, and lowermost points of the apple contour are located are the first, second, third, and fourth regions, respectively. At this time, the position of the first region relative to the conveying belt can be determined as the position of the leftmost point of the apple relative to the conveying belt, and the positions of the uppermost, rightmost, and lowermost points of the apple contour relative to the conveying belt can be determined in the same way.

[0143] In some embodiments, based on the position, the conveying speed of the conveying belt, and the kind, a mechanical arm is called to put fruits of the same kind into the same path of the conveying belt, comprising: adjusting the grabbing position of the mechanical arm to the position according to the position; determining the time when the position is conveyed to the grabbing position of the mechanical arm according to the conveying speed of the conveying belt and the shooting time of the image; controlling the mechanical arm to grab the fruit at the position according to the time; and placing the fruit into a designated path according to the kind.

[0144] The mechanical arm includes four claws respectively used for placing in four directions of up, down, left and right of the grasped object. The specified path refers to a single conveyor belt.

[0145] In an implementation, the distance of the fruit to be sorted relative to the edges of the conveyor belt is characterized as the position of the fruit to be sorted. When adjusting the grasping position of the mechanical arm, the up, down, left and right four claws of the mechanical arm are adjusted. The up claw is placed at the position of the upper edge of the fruit to be sorted relative to the edge of the conveyor belt. Similarly, the positions of the four claws are placed.

[0146] According to the conveying speed of the conveyor belt and the time when the fruit to be sorted is photographed, the time when the conveyor belt conveys the position to the position where the claws of the mechanical arm are located is determined. When the time arrives, the mechanical arm is controlled to grasp, so that the mechanical arm can grasp the fruit to be sorted. Then, according to the type of the fruit to be sorted, the specified path exclusive to the fruit to be sorted is determined.

[0147] According to the conveying speed of the conveyor belt and the photographing time of the picture, the time when the conveyor belt is in the conveying position can be obtained, so that the grasping time of the mechanical arm is adjusted. According to the position of each fruit, the mechanical arm can be adjusted to fall in the position where the fruit to be sorted will be conveyed in the future. Combining the position and the time, the mechanical arm can accurately grasp each fruit to be sorted. According to the type of the fruit, the fruit is braked into the same path, and the classification of the fruit to be sorted is completed.

[0148] In some embodiments, the type of fruit in each of the regions is obtained according to the image, including: comparing the shape, size, and skin color of the fruit in each frame with the database characteristics to determine the type of the fruit; if the matching result is multiple results, the visual sensor is called to photograph the picture of the seeds inside the fruit; and the picture characteristics of the seeds are analyzed and compared with the seed example picture corresponding to the multiple results to determine the type of the fruit.

[0149] In an implementation, image recognition technology is used to extract the shape, size and skin color characteristics of the fruit to be sorted, and the three characteristics are respectively searched in the database to match three types of search results. Then, the three types of search results are compared, and the fruit type that appears in the three types of search results at the same time is selected.

[0150] If the selected fruit type is one result, the fruit type is taken as the type of the fruit to be sorted. When the fruit type result is multiple, the visual sensor is used to further confirm the fruit to be sorted, and the nuclear magnetic resonance technology is used to image the inside of the fruit to be sorted to obtain the internal seed image of the fruit to be sorted.

[0151] Then the seed example image corresponding to the multiple categories is compared with the internal seed image of the fruit to be sorted, and the unique result category is obtained as the category of the fruit to be sorted.

[0152] The scheme analyzes the shape, size, peel color and other characteristics of the fruit to be sorted by picture analysis technology, then searches in the database according to the characteristics, and preliminarily obtains the category of the fruit to be sorted. Combining the characteristics of the fruit to be sorted, the category of the fruit to be sorted can be accurately obtained. When the category cannot be determined as one, the picture of the seed of the fruit to be sorted is compared with the seed example picture to accurately obtain the category of the fruit to be sorted, thereby improving the accuracy of obtaining the fruit to be sorted.

[0153] In some embodiments, the fruit sorting system further comprises a gravity sensing device and a price machine; the method comprises: calling the gravity sensing device to measure the weight of the fruit to be sorted; determining the specific price of the fruit to be sorted according to the weight, the price classification of the fruit to be sorted; and calling the price machine to mark the specific price of the fruit to be sorted.

[0154] Figure 4 A structural schematic diagram of a control device of a fruit sorting robot provided for an embodiment of the present application is shown in Figure 4 As shown in the figure, the control device 400 of the fruit sorting robot of the embodiment is applied to a fruit sorting system, and the fruit sorting system comprises a fruit sorting robot and a conveyor belt. The fruit sorting robot comprises a visual sensor, a near-infrared sensor and a mechanical arm. The device comprises:

[0155] The acquisition module 401 is configured to acquire an image of the fruit to be sorted on the conveyor belt by using the visual sensor;

[0156] The analysis module 402 is configured to analyze the image to determine the position and category of each fruit in the fruit to be sorted;

[0157] The classification module 403 is configured to classify fruits of the same category into the same path of the conveyor belt by using the mechanical arm based on the position, the conveying speed of the conveyor belt and the category;

[0158] The detection module 404 is configured to control the near-infrared sensor to emit near-infrared rays to the fruit to be sorted in each path according to the category;

[0159] The quality module 405 is configured to determine the quality grade of the fruit to be sorted according to the category, the emitted near-infrared rays and the received near-infrared rays for each path.

[0160] Optionally, the near-infrared sensor comprises: a plurality of light source emitters, a plurality of light source receivers; the detection module 404 controls the near-infrared sensor to emit near-infrared rays to the fruit to be sorted in each path according to the category, and is specifically used for:

[0161] According to the category, the average size, the flesh thickness, and the sugar content of the category are obtained;

[0162] According to the size, a corresponding number of light source emitters and light source receivers are started;

[0163] According to the flesh thickness and the sugar content, the detection mode of the near-infrared sensor is determined.

[0164] Optionally, when the quality module 405 determines the quality grade of the fruit to be sorted according to the category, the emitted near-infrared rays, and the received near-infrared rays, it is specifically used for:

[0165] According to the category, a preset analysis spectrum corresponding to the category is determined;

[0166] According to the emitted near-infrared rays and the received near-infrared rays, near-infrared absorption data of the fruit to be sorted is determined;

[0167] The preset analysis spectrum and the near-infrared absorption data are analyzed to determine the ingredient content of the fruit to be sorted;

[0168] According to the preset standard and the ingredient content, the quality grade of the fruit to be sorted is determined.

[0169] Optionally, when the quality module 405 determines the near-infrared absorption data of the fruit to be sorted according to the emitted near-infrared rays and the received near-infrared rays, it is specifically used for:

[0170] According to the emitted near-infrared rays, the emitted light intensity information is obtained;

[0171] According to the received near-infrared rays, the received light intensity information is obtained;

[0172] According to the emitted light intensity information and the received light intensity information, the near-infrared absorption data of the near-infrared rays is determined.

[0173] Optionally, when the analysis module 402 determines the position and category of each fruit in the fruit to be sorted according to the image, it is specifically used for:

[0174] Target recognition is performed on the image, and the conveying belt in the image is regionally divided;

[0175] The position of each region relative to the conveying belt is determined;

[0176] According to the image, the fruit category in each of the regions is obtained.

[0177] Optionally, the classification module 403 calls the mechanical arm to classify the fruits of the same category into the same path of the conveying belt based on the position, the conveying speed of the conveying belt and the category, and specifically used for:

[0178] According to the position, the grabbing position of the mechanical arm is adjusted to the position;

[0179] According to the conveying speed of the conveying belt and the shooting time of the image, the time when the position is conveyed to the grabbing position of the mechanical arm is determined;

[0180] According to the time, the mechanical arm is controlled to grab the fruit at the position;

[0181] According to the category, the fruit is placed into the designated path.

[0182] Optionally, when the analysis module 402 obtains the fruit category in each of the regions according to the image, specifically used for:

[0183] According to the shape, size and peel color of the fruit in each frame of picture, the fruit category is matched by comparing with the database features;

[0184] If the matching result is multiple results, the visual sensor is called to shoot the picture of the internal seeds of the fruit;

[0185] The picture features of the seeds are analyzed, and the fruit category is determined by comparing with the seed example picture corresponding to the multiple results.

[0186] The device of the embodiment can be used to execute the method of any of the above embodiments, and has similar implementation principles and technical effects, which will not be described here.

[0187] Figure 5 A structural schematic diagram of an electronic device provided by an embodiment of the present application is shown in FIG. 5, and the electronic device 500 of the embodiment can include a memory 501 and a processor 502. Figure 5

[0188] The memory 501 stores a computer program capable of being loaded and executed by the processor 502 to execute the method in the above embodiments.

[0189] The processor 502 and the memory 501 are connected, for example, through a bus.

[0190] Optionally, the electronic device 500 can further include a transceiver. It should be noted that the transceiver in actual application is not limited to one, and the structure of the electronic device 500 does not constitute a limitation on the embodiments of the present application.​

[0191] The processor 502 can be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other programmable logic device, transistor logic device, hardware component or any combination thereof. It can implement or execute the various exemplary logical blocks, modules and circuits described in connection with the disclosure. The processor 502 can also be a combination of computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0192] The bus can include a path for transmitting information between the above-mentioned components. The bus can be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is shown in the figure, but it does not mean that there is only one bus or only one type of bus.

[0193] The memory 501 can be a ROM (Read Only Memory) or other type of static storage device that can store static information and instructions, a RAM (Random Access Memory) or other type of dynamic storage device that can store information and instructions, an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, an optical disk storage (including a compact disk, a laser disk, an optical disk, a digital versatile disk, a Blu-ray disk, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and capable of being accessed by a computer, but not limited thereto.

[0194] The memory 501 is configured to store application program codes for implementing the solutions of the present application, and the processor 502 is configured to control the execution of the application program codes. The processor 502 is configured to execute the application program codes stored in the memory 501 to implement the above-mentioned contents shown in the method embodiments.

[0195] The electronic device includes, but is not limited to, a mobile terminal such as a mobile phone, a notebook computer, a digital broadcast receiver, a PDA (Personal Digital Assistant), a PAD (Tablet Personal Computer), a PMP (Portable Multimedia Player), a car terminal (for example, a car navigation terminal), and the like, and a fixed terminal such as a digital TV, a desktop computer, and the like. It can also be a server or the like. Figure 5 The electronic device shown is only an example, and should not bring any limitation to the functions and use range of the embodiments of the present application.

[0196] The electronic device of the present embodiment can be used to execute the method of any one of the above embodiments, and has similar implementation principles and technical effects, which will not be described here.

[0197] The present application also provides a computer readable storage medium, which stores a computer program capable of being loaded and executed by a processor to perform the method in the above embodiments.

[0198] Those skilled in the art can understand that all or part of the steps of the above-mentioned method embodiments can be completed by program instruction related hardware. The above-mentioned program can be stored in a computer readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the above-mentioned storage medium includes ROM, RAM, magnetic disk or optical disk and various media that can store program codes.

Claims

1. A control method of a fruit picking robot, characterized by, The application is applied to a fruit sorting system, and the fruit sorting system comprises a fruit sorting robot, a conveyor belt, the fruit sorting robot comprising a visual sensor, a near-infrared sensor and a mechanical arm; The conveyor belt is used for conveying fruits to be sorted; The fruit sorting robot is used for collecting images of the fruits to be sorted on the conveyor belt by using the visual sensor; analyzing the images to determine the position and type of each fruit to be sorted; calling the mechanical arm to group fruits of the same type in the same path of the conveyor belt based on the position, the conveying speed of the conveyor belt and the type; controlling the near-infrared sensor to emit near-infrared rays to the fruits to be sorted in each path according to the type; and determining the quality grade of the fruits to be sorted according to the type, the emitted near-infrared rays and the received near-infrared rays for each path. The method comprises: Collecting images of the fruits to be sorted on the conveyor belt by using the visual sensor; Analyzing the images to determine the position and type of each fruit to be sorted; Calling the mechanical arm to group fruits of the same type in the same path of the conveyor belt based on the position, the conveying speed of the conveyor belt and the type; Controlling the near-infrared sensor to emit near-infrared rays to the fruits to be sorted in each path according to the type; Determining the quality grade of the fruits to be sorted according to the type, the emitted near-infrared rays and the received near-infrared rays for each path. The near-infrared sensor comprises a plurality of light source emitters and a plurality of light source receivers. The controlling the near-infrared sensor to emit near-infrared rays to the fruits to be sorted in each path according to the type comprises: Obtaining the average size, flesh thickness and sugar content of the type according to the type; Starting a corresponding number of the light source emitters and the light source receivers according to the size; Determining the detection mode of the near-infrared sensor according to the flesh thickness and the sugar content, wherein the detection mode comprises reflection detection and transmission detection; The starting a corresponding number of the light source emitters and the light source receivers according to the size comprises: determining a preset standard gradient corresponding to the size, each preset standard gradient corresponding to a number; and starting the light source emitters and the light source receivers corresponding to the number corresponding to the preset standard gradient corresponding to the size. The determining the quality grade of the fruits to be sorted according to the type, the emitted near-infrared rays and the received near-infrared rays for each path comprises: Determining a preset analysis spectrum corresponding to the type according to the type; Determining near-infrared absorption data of the fruits to be sorted according to the emitted near-infrared rays and the received near-infrared rays; Analyzing the preset analysis spectrum and the near-infrared absorption data to determine the ingredient content of the fruits to be sorted; Determining the quality grade of the fruits to be sorted according to a preset standard and the ingredient content.

2. The method of claim 1, wherein, The determining the near-infrared absorption data of the fruits to be sorted according to the emitted near-infrared rays and the received near-infrared rays comprises: Obtaining emitted light intensity information according to the emitted near-infrared rays; Obtaining received light intensity information according to the received near-infrared rays. Determine near-infrared absorption data of the near-infrared rays according to the emitted light intensity information and the received light intensity information.

3. The method of claim 1, wherein, The determining the position and the kind of each fruit in the fruit to be sorted according to the image comprises: Performing target recognition on the image, and dividing the conveying belt in the image into regions; Determining the position of each region relative to the conveying belt; Obtaining the kind of fruit in each region according to the image.

4. The method of claim 3, wherein, The calling the mechanical arm to classify fruits of the same kind into the same path of the conveying belt based on the position, the conveying speed of the conveying belt and the kind comprises: Adjusting the grabbing position of the mechanical arm to the position according to the position; Determining the time when the position is conveyed to the grabbing position of the mechanical arm according to the conveying speed of the conveying belt and the shooting time of the image; Controlling the mechanical arm to grab the fruit at the position according to the time; Placing the fruit into a designated path according to the kind.

5. The method of claim 3, wherein, The obtaining the kind of fruit in each region according to the image comprises: Comparing the shape, size and peel color of the fruit in each frame with the characteristics in a database to match the kind of the fruit; If there are multiple matching results, calling the visual sensor to shoot a picture of the seeds inside the fruit; Comparing the picture characteristics of the seeds with seed example pictures corresponding to the multiple results to determine the kind of the fruit.

6. A control device for a fruit picking robot, characterized in that The fruit sorting system comprises a fruit sorting robot and a conveying belt. The fruit sorting robot comprises a visual sensor, a near-infrared sensor and a mechanical arm. The fruit sorting robot is configured to collect an image of the fruit to be sorted on the conveying belt by using the visual sensor, analyze the image to determine the position and kind of each fruit in the fruit to be sorted, call the mechanical arm to classify fruits of the same kind into the same path of the conveying belt based on the position, the conveying speed of the conveying belt and the kind, control the near-infrared sensor to emit near-infrared rays to the fruit to be sorted in each path according to the kind, and determine the quality grade of the fruit to be sorted according to the kind, the emitted near-infrared rays and the received near-infrared rays for each path. The device comprises: A collection module configured to collect an image of the fruit to be sorted on the conveying belt by using the visual sensor; An analysis module configured to analyze the image to determine the position and kind of each fruit in the fruit to be sorted; A classification module configured to call the mechanical arm to classify fruits of the same kind into the same path of the conveying belt based on the position, the conveying speed of the conveying belt and the kind; A detection module configured to control the near-infrared sensor to emit near-infrared rays to the fruit to be sorted in each path according to the kind; A quality module configured to determine the quality grade of the fruit to be sorted according to the kind, the emitted near-infrared rays and the received near-infrared rays for each path. The near-infrared sensor comprises a plurality of light source emitters and a plurality of light source receivers. The detection module, when controlling the near-infrared sensor to emit near-infrared rays to the fruit to be sorted in each path according to the category, is specifically configured to: According to the category, obtain the average size, flesh thickness, and sugar content of the category; According to the size, start a corresponding number of light source emitters and light source receivers; According to the flesh thickness and the sugar content, determine a detection mode of the near-infrared sensor, the detection mode including reflection detection and transmission detection; The detection module, when starting a corresponding number of light source emitters and light source receivers according to the size, is specifically configured to: determine a preset standard gradient corresponding to the size, each preset standard gradient corresponding to a number; and start the light source emitters and light source receivers corresponding to the number corresponding to the preset standard gradient corresponding to the size; The quality module, when determining the quality grade of the fruit to be sorted according to the category, the emitted near-infrared rays, and the received near-infrared rays, is specifically configured to: According to the category, determine a preset analysis spectrum corresponding to the category; According to the emitted near-infrared rays and the received near-infrared rays, determine near-infrared absorption data of the fruit to be sorted; Analyze the preset analysis spectrum and the near-infrared absorption data to determine the ingredient content of the fruit to be sorted; According to a preset standard and the ingredient content, determine the quality grade of the fruit to be sorted.

7. An electronic device, comprising: Comprise: A memory and a processor; The memory is configured to store program instructions; The processor is configured to call and execute the program instructions in the memory to execute the control method of the fruit sorting robot according to any one of claims 1-5.

Images