CN115015495A - Dynamic close-range miniature intelligent sensor for quality of growing spherical fruits and vegetables - Google Patents
Dynamic close-range miniature intelligent sensor for quality of growing spherical fruits and vegetables Download PDFInfo
- Publication number
- CN115015495A CN115015495A CN202210619329.0A CN202210619329A CN115015495A CN 115015495 A CN115015495 A CN 115015495A CN 202210619329 A CN202210619329 A CN 202210619329A CN 115015495 A CN115015495 A CN 115015495A
- Authority
- CN
- China
- Prior art keywords
- fruit
- quality
- module
- detection
- sensor
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/02—Food
Abstract
The utility model provides a miniature intelligent perception ware of spherical fruit vegetables quality developments closely in growing, relates to the non-destructive testing field of agricultural and animal products, provides a miniature intelligent perception ware of spherical fruit vegetables quality developments closely in growing, including target identification module, flexible control module, quality detection module, target identification module includes degree of depth camera and range finding sensor, quality detection module includes the trigger, the spatial position of target fruit is confirmed based on image processing technique to target identification module, guides the path planning of perception ware, based on the flexible control module's of target identification result control flexible. And the target recognition module obtains the actual size of the fruit diameter and the actual distance of the equator position of the fruit relative to the camera through a correction algorithm. Three micro-motion triggers are uniformly arranged on the circumference of the quality information acquisition module to detect pressure and judge the joint tightness during acquisition. The fruit detection device is used for fruit detection, reduces labor intensity compared with manual detection, and reduces volume and development cost compared with mechanical arms.
Description
Technical Field
The invention relates to the field of nondestructive testing of agricultural and livestock products, in particular to a dynamic short-distance miniature intelligent sensor for the quality of growing spherical fruits and vegetables.
Background
The growth of fruit is a complex and time-consuming process, and generally undergoes a seedling stage, a growing stage, a flowering stage, a fruit drop stage, a fruit expansion stage and a fruit maturation stage. Fruit growers need to complete weeding, watering and fertilizing, pesticide spraying and pruning and other works in the orchard in different growth periods, so that the fruit yield and the fruit quality are directly influenced by the supervision level of the orchard. With the increase of labor cost, the mechanized production of the orchard is very important, and agricultural intelligent equipment is an important factor for reducing labor and cost. The intelligent management of fruit production needs the scale and standardization of planting, and the scale planting can improve the utilization rate of a machine, and the standardized planting can exert the efficacy of the machine.
The image technology or remote sensing technology collects earth surface information from the height of several meters to several hundred meters, and is a common orchard monitoring method at present. However, the method has certain limitations, is greatly influenced by environments such as weather, illumination and the like, has a long detection distance, is limited in detection resolution and precision, and cannot accurately detect the quality information of the fruits. The accurate acquisition of the growth information of each fruit tree is very necessary for accurate management, a sensor with small volume and high detection accuracy is required to acquire characteristic information in a short distance, and the research and the invention are not provided at present.
Disclosure of Invention
Therefore, the invention provides a dynamic short-distance miniature intelligent sensor for the quality of growing spherical fruits and vegetables, and aims to accurately acquire the quality information of each fruit tree.
In order to achieve the above purposes, the technical scheme adopted by the invention is as follows:
a dynamic close-range miniature intelligent sensor for the quality of growing spherical fruits and vegetables comprises a target identification module, a telescopic control module and a quality information acquisition module;
the target identification module is used for identifying a target and obtaining a space coordinate of a sensor and the target;
the telescopic control module is used for adjusting the position of the quality information acquisition module;
the quality information acquisition module is used for detecting the quality of the fruits;
the target recognition module comprises a depth camera and a ranging sensor, and is used for remotely recognizing a target based on a deep learning algorithm and controlling a sensor to approach to a fruit; the target recognition module obtains the actual distance between the sensor and the fruit and the actual size of the fruit based on a correction algorithm, and controls the sensor to reach the detection position; the target identification module is based on the distance measuring sensor detects the distance between the quality information acquisition module and the fruit, and controls the telescopic control module to enable the quality information acquisition module to be attached to the fruit.
Further, the correction algorithm comprises the steps of:
s1, correcting the fruit diameter size in the image to the actual size;
s2, calculating the elevation angle and the actual distance of the fruit equator position relative to the camera;
s3, finally, calculating the spatial position of the fruit relative to the detection module through coordinate translation; wherein, the correction formula of step S1 is as follows:
R＝l 2 A
wherein R is the actual fruit radius l 2 Is the diameter of the section of the fruit which can be observed by the visual field of the camera, A is a correction coefficient and is calculated by the following formula:
wherein l 3 For fruit with high stature, | 4 The distance from the lower edge of the fruit image to the horizontal line, and f is the fruit image distance;
the calculation formula of step S2 is as follows:
elevation angle:
distance: l 6
Wherein l 1 Horizontal distance of the fruit edge from the camera.
Further, the quality information acquisition module comprises a spectrum sensor, an annular light source, a shading shell and a trigger;
the annular light source is used for actively emitting light;
the spectrum sensor is used for collecting spectrum information diffused and reflected inside the fruit;
the shading shell is used for isolating stray light of the external environment;
the quantity of the triggers is at least 3, the triggers are uniformly distributed along the circumference of the quality information acquisition module and are used for judging the joint tightness of the quality information acquisition module and the fruit.
Furthermore, the trigger comprises a probe, a sleeve, a first spring, a base, an inductive contact, a second spring and a stop lever; the probe is of a hollow structure at the end with the larger diameter, the spring I is arranged in the probe, the probe is provided with a transverse through hole, the stop lever penetrates through the transverse through hole to stop the spring I in the probe, and the diameter range of the end with the smaller diameter of the probe is 1mm to 2 mm; the base is of a hollow structure, a second spring is arranged in the base, and two sensing contacts which are symmetrically distributed are arranged on the wall of the base; the probe is arranged in the base hole and clamped in the base by the sleeve; the stop lever and the inductive contact are made of conductive materials; the stiffness of the second spring is smaller than that of the first spring.
Furthermore, the telescopic control module comprises a front support, a rear support, a sliding rail, a sliding block, a sleeve, a push rod, a connecting seat, a gear, a rack and a speed reducing motor, wherein the speed reducing motor drives the gear, the gear drives the rack, the rack is fixed on the sliding block, the sleeve is integrally arranged on the sliding block, the sleeve is of a hollow structure, the push rod and a push rod buffer spring are movably arranged in the sleeve, and one end of the push rod is connected with the quality information acquisition module through the connecting seat; the sliding block is movably sleeved on the sliding rail through a through hole, and two ends of the sliding rail are respectively fixed on the front support and the rear support.
Furthermore, the slide rail is made of carbon fiber materials, and the front support, the rear support, the sliding block, the sleeve, the push rod and the connecting seat are made of synthetic resin materials.
The method for judging the validity of the collected data based on the growing spherical fruit and vegetable quality dynamic close-range micro intelligent sensor comprises the following steps:
the method comprises the following steps: waiting for detection;
step two: judging whether the trigger is triggered or not, and if not, returning to the first step;
step three: judging whether the measured spectrum is in a limited interval or not, and returning to the first step if the measured spectrum is not in the limited interval;
step four: storing the data;
step five: judging whether to continue detection, if so, returning to the first step;
step six: and (6) ending.
The control method of the telescopic control module based on the growing spherical fruit and vegetable quality dynamic close-range micro intelligent sensor comprises the following steps:
the method comprises the following steps: waiting for detection;
step two: judging whether the detection distance is triggered or not, and if not, returning to the first step;
step three: PID adjusts the rotating speed of the speed reducing motor;
step four: the push rod extends out;
step five: judging whether information is acquired or not, and returning to the third step if the information is not acquired;
step six: retracting the push rod;
step seven: judging whether to continue detection, if so, returning to the first step;
step eight: and (6) ending.
The dynamic close-range miniature intelligent sensor for the quality of growing spherical fruits and vegetables has the following beneficial effects: the trigger is adopted to detect the attachment of the fruits, so that the reliability of the acquired information is ensured; the correction algorithm of the fruit size and the actual detection points is provided, and the spatial position and the detection points of the fruit can be accurately judged; the sensor is used for fruit detection, reduces the labor intensity compared with manual detection, can detect fruits at higher positions, ensures the safety, and reduces the volume and the development cost compared with mechanical arms.
Drawings
The invention has the following drawings:
FIG. 1 is a diagram of a sensor architecture according to the present invention;
FIG. 2 is a schematic diagram of an application scenario of the sensor of the present invention
FIG. 3 is a schematic diagram of the geometric principle of the correction algorithm of the present invention;
FIG. 4 is a block diagram of a quality information acquisition module of the present invention;
FIG. 5 is a block diagram of the trigger of the present invention;
FIG. 6 is a flow chart of the present invention for determining validity of collected data;
FIG. 7 is a control flow diagram of the telescoping control module of the present invention.
Reference numerals:
1. trigger, 2. shading shell, 3. sponge, 4. connecting seat, 5. push rod, 6. frame, 71, 72: slider, 81, 82: the device comprises a sleeve, 9 parts of a rear support, 10 parts of a rack, 11 parts of a speed reducing motor, 12 parts of a gear, 13 parts of a sliding rail, 14 parts of a front support, 15 parts of a depth camera, 16 parts of a distance measuring sensor, 17 parts of a spectrum sensor, 18 parts of an annular LED, 101 parts of a light source probe, 102 parts of the sleeve, 103 parts of a first spring, 104 parts of a base, 105 parts of a sensing contact, 106 parts of a second spring, 107 parts of a stop lever, 108 parts of a through hole
Detailed Description
The present invention is described in further detail below with reference to figures 1-7.
Fig. 2 shows a schematic view of an application scene of the sensor of the invention, and the invention provides a dynamic short-distance micro intelligent sensor for the quality of growing spherical fruits and vegetables, which is arranged on a micro intelligent unmanned aerial vehicle, and collects the quality characteristic information of the fruits and detects the internal and external quality of the fruits in a dynamic hovering posture in the process of being close to the fruits on a tree in a short distance. The sensor comprises a target identification module, a quality information acquisition module and a telescopic control module. The target identification module searches and tracks the position of the fruit in real time, and can automatically correct the size of the fruit and the coordinates of the actual detection point at different distances and angles. The telescopic control module is used for controlling the movement of the quality information acquisition module in the horizontal direction, and automatically extends out and retracts according to the identification result, so that the information acquisition distance is prolonged by a small volume. A judgment mode based on serial connection of a trigger and a spectrum verification result is provided, and whether a sensing quality information acquisition module is attached to a fruit or not is sensed. The invention can be used for monitoring the internal and external quality, the nutrition deficiency, the maturity condition and the like of the fruit, establishes a quality information dynamic model in the fruit growth by monitoring a plurality of quality state parameters in the fruit growth in real time, realizes the digital management, the quality detection and the Internet of things control in the fruit growth, achieves the purpose of high-quality and high-efficiency production, provides important big data information for the intelligent decision of an unmanned orchard, improves the information collection, the intelligent decision and the accurate operation capability of the fruit production, and is beneficial to promoting the formation of an information intelligent integral solution for the production of bulk agricultural products.
As shown in FIG. 1, the structure diagram of the growing spherical fruit and vegetable quality dynamic close-range micro intelligent sensor comprises a target identification module, a telescopic control module and a quality information acquisition module. The target identification module comprises a depth camera and a ranging sensor; the telescopic control module comprises a front bracket 14, a rear bracket 9, a slide rail 13, slide blocks 71 and 72, sleeves 81 and 82, a push rod 5, a connecting seat 4, a gear 12, a rack 10 and a speed reduction motor 11; the quality information acquisition module comprises a flexible sponge 3, a shading shell 2 and a trigger 1. Wherein the depth camera 15, the range sensor 16, the gear motor 11, the trigger 1 and the control of the detection can be controlled by a micro processor such as a raspberry pi.
The object recognition module includes a depth camera 15 and a range sensor 16. Wherein the depth camera 15 is fixed on the frame as a long-distance recognition means, and the distance measuring sensor 16 is installed in the light shielding shell 2 as a short-distance recognition means.
The target identification module firstly identifies a target at a long distance by using a deep learning algorithm Yolov5 to obtain coordinate information of a fruit in an image, and then maps the coordinate information to the depth information to obtain original identification results X, Y and Z. And then comprehensively judging and deciding based on the detection result and the depth information of the Yolov5, and screening out fruits which are not blocked by branches and leaves at the current position of the sensor, so as to facilitate subsequent detection. Then, detection is carried out on a specific fruit, the sensor is gradually close to the fruit in the detection process, the camera base line and the sensor are not at the same horizontal position, the position of the specific fruit relative to the sensor needs to be corrected in the approaching process, and therefore a correction algorithm for the size of the fruit and the detection point is introduced. It should be noted that this process is a far-to-near identification, and the drone gradually flies through. The distance measuring sensor is arranged at the center position of the quality information acquisition module, can be over fruit when acquiring information, and aims to ensure that the shading shell covers the fruit surface when acquiring information is gradually pressed close to the fruit surface, so that the fruit target is lost when a user looks at the image, and the distance measuring sensor can judge the real-time distance of the fruit through the distance at the moment, thereby providing distance guide for the telescopic control module. It should be noted that this process is a process from close to close, the drone is hovering, and close to close is completed by the push rod until close.
The correction algorithm is used for eliminating distortion errors existing when the camera base line and the fruit are not at the same horizontal height, and is used for accurately judging the spatial position and the detection point of the fruit. Firstly, correcting the fruit diameter size preliminarily obtained in an image into an actual size; and then calculating the actual detection point, namely the actual distance of the equator position relative to the camera according to the elevation angle, and finally converting the spatial position of the fruit relative to the detection module through coordinate translation.
As shown in fig. 3, taking the projection in the X-axis direction as an example, point O is the center of a circle of a fruit, point E is the position of a depth camera, point J is the position of a detection module, and point F is the imaging position of the camera. According to the principle of pinhole imaging, the projection position of the fruit on the image is GI, the real reflection area of GI is AB length, and the projection position is influenced by the visual field area of the camera and the pinhole imaging, the length of AB can be observed only by the spherical fruit at the position of the camera, but the AB cannot reflect the real diameter of the circular fruit. The calibration process is as follows:
the information obtained by the depth camera can BE used for knowing the distances AE and BE of the outer edges of the fruits, the detection diameter AB, the projection length GI of the fruits in the image, the distance FG from the lower edge of the fruit image to the horizontal line and the image distance EF, and for convenience of description, AE is BE and L 1 ，AB＝l 2 ，GI＝l 3 ，FG＝l 4 ，EF＝f。
Wherein, the actual radius length is R, and the correction coefficient is A.
Different from picking collarThe field concerns the outline of the circumscribed circle of an apple and the quality information acquisition of the invention requires the alignment of the acquisition module around the equator of the apple. When an included angle exists between the axis of the quality detection module and the camera base line, the center position of an image acquired by the camera is actually the position D of the fruit, and the position needing to be detected is located at the position C of the equator of the fruit, so that the coordinates of the detected fruit center need to be further corrected. In Δ OCE, OC ═ R and OE ═ l 5 ，CE＝l 6 ；
It can be obtained that the elevation angle of the detecting point relative to the camera is
A distance of l 6 。
Finally, the spatial position relationship between the detection point and the quality detection module needs to be calculated, and can be obtained by converting a translation coordinate system:
P′＝P+ΔP (10)
wherein, P' is the coordinate of the point C relative to the quality detection module, P is the coordinate of the point C relative to the depth camera, and Δ P is the translation transformation matrix.
The invention uses a detection mode of a telescopic sensor, the sensor can actively extend out and retract from the side, then the quality of the fruit growing on the tree is detected in a hovering posture, and the hovering action and the collecting action are separated by the detection mode, so that the stability of the dynamic detection process is ensured. The detection mode is specifically implemented by using a telescopic control module to prolong the acquisition distance and controlling the telescopic operation according to the target identification result. The control flow of the telescoping control module is shown in fig. 7.
Regarding the telescopic control, the invention provides a detection mode of a telescopic sensor, which can be close to fruits on a tree in a short distance and detect the internal and external quality of the fruits in a dynamic hovering posture. The telescopic control module is used for controlling the movement of the quality information acquisition module in the horizontal direction, automatically extending out and retracting according to the identification result and prolonging the information acquisition distance.
As shown in fig. 1, the telescopic control module is mounted on the frame 6, and has four threaded holes fixed to the frame 6 and located at the bottom of the front support 14 and the bottom of the rear support 9, respectively. The module further comprises a slide rail 13, sleeves 81, 82, sliders 71, 72, a push rod 5, a connecting seat 4, a gear 12, a rack 10 and a reduction motor 11.
In order to control the extension and retraction of the quality detection module, the telescopic control module uses a gear rack transmission mechanism, a speed reduction direct current motor 11 provides power, and a gear 12 is driven by the speed reduction motor to drive a rack 10 to move.
The rack 10 is fixed on the sliding blocks 71, 72, and in order to satisfy the smooth sliding of the sliding blocks, the sliding blocks 71, 72 are provided with two through holes which are matched with the sliding rails 13, so that the sliding blocks are limited to move back and forth smoothly. Two ends of the slide rail 13 are respectively connected and fixed with the front bracket 14 and the rear bracket 9.
The sleeves 81 and 82 are fixed relative to the sliding blocks 71 and 72, the sleeves 81 and 82 are used for loading the push rod 5, and springs are arranged inside the sleeves 81 and 82 and used for providing an axial buffer area, so that the light shielding shell 2 can be better attached to fruits, and trace axial displacement can be provided during information acquisition.
The telescopic control module is always in a retraction state in the process of searching a detection target by the sensor, the push rod can be extended out when the detection action is executed, and the push rod retracts again after the detection action is finished. Wherein the extending and retracting actions are controlled by the forward and reverse rotation of the speed reducing motor. The telescopic module has two working modes, the material is carbon fiber and resin, and the weight is not more than 50 g.
Referring to fig. 4, the quality information acquisition module includes a spectrum sensor 17, an annular LED light source 18, a light shielding housing 2 and a trigger 1. The annular LED light source 18 is used for actively emitting light, the light is subjected to diffuse reflection inside the fruit, the light intensity of the diffuse reflection carries the internal information of the fruit, and the spectrum sensor 17 collects the light intensity of the diffuse reflection to obtain the spectrum information. The light shielding shell 2 is used for shielding stray light of the external environment. The module is controlled by a single chip microcomputer, IIC communication is adopted, and collection is triggered when the close fitting is judged. And the acquired spectral information is brought into a prediction model to obtain a detection result. The spectral information and detection results may then be sent to a remote terminal using Wi-Fi or uploaded to a server. In order to sense and verify the reliability of the acquired information, a sensing mode based on serial connection of a trigger and spectrum verification is provided for judging whether the acquired data is valid or not. Firstly, three micro-motion triggers are uniformly arranged on the circumference of the quality information acquisition module to detect pressure to judge the joint tightness during acquisition, and then, whether the acquired information is effective is checked and checked on data.
Different from a conventional detection module, the invention adds the trigger in the shading shell under the condition of not increasing extra volume, three thin probes in the shading shell can be detected and attached, the thin probes occupy little internal volume, the influence of the light path of light refraction and reflection in the shading cover is avoided as far as possible, and the diameter range of the extending end of the probe is 1 mm-2 mm.
As shown in fig. 5, the trigger includes a probe 101, a sleeve 102, a first spring 103, a base 104, a sensing contact 105, a second spring 106, and a stop lever 107. The probe 101 is internally provided with a hole, a first spring 103 is arranged, the probe 101 is provided with a through hole 108, a stop rod 107 penetrates through the through hole 108, and the first spring 103 is clamped inside the probe by the stop rod 107. The base 104 is provided with a second spring 106 through an inner opening, two sensing contacts 105 are symmetrically distributed on the wall of the base, and the probe 101 is arranged in the hole of the base 104 and clamped in the base 104 by the sleeve 102.
The stop lever 107 and the inductive contact 105 are conductive copper sheets. Spring two 106 is softer than spring one 103. When the probe 101 is pressed by the axial force due to the reverse acting force, the second spring 106 is compressed first, the stop lever 107 is pressed on the inductive contact 105, and the current of the two inductive contacts is conducted at the moment, so that the detection is in place. If the probe continues to be subjected to the reverse acting force, the first spring 103 is compressed, and the probe 101 can continue to axially move at the moment, wherein the moving interval is the length of the through hole 108, and the moving interval provides an axial buffer area for the probe, so that the trigger disconnection caused by the shaking in the detection process is prevented.
The trigger 1 is mounted on the housing of the light shielding housing 2, penetrates through the bottom of the housing, and is fixed at the bottom of the light shielding housing 2. The top of the probe 101 is a cylinder with a diameter of 1mm, and the probe 101 is located inside the light-shielding housing. The combination of 3 triggers at least can effectively detect the laminating degree of spherical fruit and shading shell, indicates that the laminating state is good when the trigger is all triggered and does not have the light leak, if the trigger does not all trigger and indicates that the laminating is not sealed, need gather spectral information again.
Aiming at the dynamic detection environment, the invention can automatically judge after collecting the spectrum information. The quality detection module designed by the invention can effectively detect the validity of data, only when the spectrum is in a limited interval and 3 triggers are met, the data is judged to be valid, the current data is stored, otherwise, the data is acquired again, and the judgment process is as shown in fig. 6. The spectral interval is an empirical value and can be set for a specific use case. It should be noted that this process is only used as a process for collecting the final data validity judgment, and data should be checked during the information collection process. For example, the continuous collection is carried out for multiple times, the distribution condition of the samples is counted, abnormal values are removed, then the average is carried out, and the method also belongs to the effective gain of the dynamic collection of the sensor. The elimination of abnormal values is performed by performing cluster analysis after repeated collection, and then eliminating abnormal values with small probability.
The invention uses microcontroller as real-time control system, and its hardware can be raspberry pi or micro host. The microcontroller controls different tasks in a multithreading mode, integrates devices such as a distance measuring sensor, a depth camera, a trigger and a multispectral sensor in various communication modes, and simultaneously gives consideration to wireless transmission and data uploading of data. The wireless transmission meaning of the data is that the data can interact with a ground station in real time, not only can work independently, but also can be controlled by the ground station to transmit real-time instructions, images, data, detection results and the like. The significance of uploading data is that big data can be obtained, the internal and external quality is detected by combining image and spectrum means, the growth situation of the whole park can be effectively obtained, information is uploaded in real time based on the internet of things technology, and the decision of an expert system is facilitated.
Aiming at the growth state of the fruit trees, the method can accurately predict a plurality of growth parameters, such as a plurality of indexes of fruit shape, fruit diameter, hardness, acidity, sugar degree, moisture content, starch content and the like in the growth process, can comprehensively decide and guide the management of the orchard according to the parameters, and accurately monitor the quality, nutrition and maturity of each fruit tree and predict the optimal picking period. Compared with the traditional working mode, the orchard monitoring system has the advantages that effective growth data are extracted at different growth stages according to specific functions, and the orchard monitoring problem can be accurately solved.
The monitoring of the fruit trees can establish a dynamic perception model, deeply excavate big data information, and count the daily growth condition and fruit nutrition condition of the orchard. The method can not only obtain the growth trend of the current orchard, but also predict the growth trend of the next step according to the past collected information and the current information, so that the intelligent management early warning of the orchard is realized, and the prediction is provided for the mature picking of fruits.
In addition, the detection data are uploaded to a cloud server for maintenance through an uploading module of the controller, massive data are managed on the cloud server, and key information such as the production place, the variety, the tree age and the like is marked, so that model maintenance is achieved. And extracting corresponding data for modeling aiming at a specific variety, and updating the models aiming at the models in different years. Under the support of big data, the model of the sensor can be updated and maintained according to the uploading integration of the data.
Those not described in detail in this specification are within the skill of the art.
Claims (8)
1. The utility model provides a miniature intelligent perceptron of spherical fruit vegetables quality developments closely in growing which characterized in that includes:
the system comprises a target identification module, a telescopic control module and a quality information acquisition module;
the target identification module is used for identifying a target and obtaining a space coordinate of a sensor and the target;
the telescopic control module is used for adjusting the position of the quality information acquisition module;
the quality information acquisition module is used for detecting the quality of the fruits;
the target recognition module comprises a depth camera and a ranging sensor, and is used for remotely recognizing a target based on a deep learning algorithm and controlling a sensor to approach to a fruit; the target recognition module obtains the actual distance between the sensor and the fruit and the actual size of the fruit based on a correction algorithm, and controls the sensor to reach the detection position; the target identification module is based on the distance measuring sensor detects the distance between the quality information acquisition module and the fruit, and controls the telescopic control module to enable the quality information acquisition module to be attached to the fruit.
2. The growing spherical fruit and vegetable quality dynamic near-field micro intelligent sensor according to claim 1, wherein the correction algorithm comprises the steps of:
s1, correcting the fruit diameter size in the image to be the actual size;
s2, calculating the elevation angle and the actual distance of the fruit equator position relative to the camera;
s3, finally, calculating the spatial position of the fruit relative to the detection module through coordinate translation;
wherein:
step S1 has the following correction formula:
R＝l 2 A
wherein R is the actual fruit radius l 2 Is the diameter of the fruit section observed by the camera view, A is the correction coefficient, and A is calculated by the following formula:
wherein l 3 For fruit with high stature, | 4 The distance from the lower edge of the fruit image to the horizontal line, and f is the fruit image distance;
the calculation formula of step S2 is as follows:
elevation angle:
distance: l. the 6
Wherein l 1 Horizontal distance of the fruit edge from the camera.
3. The growing spherical fruit and vegetable quality dynamic close-range micro intelligent sensor according to claim 2, wherein:
the quality information acquisition module comprises a spectrum sensor, an annular light source, a shading shell and a trigger;
the annular light source is used for actively emitting light;
the spectrum sensor is used for collecting spectrum information diffused and reflected inside the fruit;
the shading shell is used for isolating stray light of the external environment;
the quantity of the triggers is at least 3, the triggers are uniformly distributed along the circumference of the quality information acquisition module and are used for judging the joint tightness of the quality information acquisition module and the fruit.
4. The growing spherical fruit and vegetable quality dynamic close-range micro intelligent sensor according to claim 3, wherein: the trigger comprises a probe, a sleeve, a first spring, a base, an induction contact, a second spring and a stop lever; the probe is of a hollow structure at the end with the larger diameter, the spring I is arranged in the probe, the probe is provided with a transverse through hole, the stop lever penetrates through the transverse through hole to stop the spring I in the probe, and the diameter range of the end with the smaller diameter of the probe is 1mm to 2 mm; the base is of a hollow structure, a second spring is arranged in the base, and two sensing contacts which are symmetrically distributed are arranged on the wall of the base; the probe is arranged in the base hole and clamped in the base by the sleeve; the stop lever and the inductive contact are made of conductive materials; the stiffness of the second spring is smaller than that of the first spring.
5. The growing spherical fruit and vegetable quality dynamic close-range micro intelligent sensor according to claim 4, wherein: the telescopic control module comprises a front support, a rear support, a slide rail, a slide block, a sleeve, a push rod, a connecting seat, a gear, a rack and a speed reducing motor, wherein the speed reducing motor drives the gear, the gear drives the rack, the rack is fixed on the slide block, the sleeve is integrally arranged on the slide block, the sleeve is of a hollow structure, the push rod and a push rod buffer spring are movably arranged in the sleeve, and one end of the push rod is connected with a quality information acquisition module through the connecting seat; the sliding block is movably sleeved on the sliding rail through a through hole, and two ends of the sliding rail are respectively fixed on the front support and the rear support.
6. The growing spherical fruit and vegetable quality dynamic close-range micro intelligent sensor according to claim 5, wherein: the slide rail is made of carbon fiber materials, and the front support, the rear support, the slide block, the sleeve, the push rod and the connecting seat are made of synthetic resin materials.
7. The method for judging the validity of the collected data of the growing spherical fruit and vegetable quality dynamic close-range micro intelligent sensor based on claim 4 comprises the following steps:
the method comprises the following steps: waiting for detection;
step two: judging whether the trigger is triggered or not, and if not, returning to the first step;
step three: judging whether the measured spectrum is in a limited interval or not, and if not, returning to the first step;
step four: storing the data;
step five: judging whether to continue detection, if so, returning to the first step;
step six: and (6) ending.
8. The control method of the expansion control module of the growing spherical fruit and vegetable quality dynamic close-range micro intelligent sensor based on the claim 5 comprises the following steps:
the method comprises the following steps: waiting for detection;
step two: judging whether the detection distance is triggered or not, and if not, returning to the first step;
step three: PID adjusts the rotating speed of the speed reducing motor;
step four: the push rod extends out;
step five: judging whether information is acquired or not, and returning to the third step if the information is not acquired;
step six: retracting the push rod;
step seven: judging whether to continue detection, if so, returning to the first step;
step eight: and (6) ending.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210619329.0A CN115015495B (en) | 2022-06-02 | 2022-06-02 | Dynamic close-range miniature intelligent sensor for quality of growing spherical fruits and vegetables |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210619329.0A CN115015495B (en) | 2022-06-02 | 2022-06-02 | Dynamic close-range miniature intelligent sensor for quality of growing spherical fruits and vegetables |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115015495A true CN115015495A (en) | 2022-09-06 |
| CN115015495B CN115015495B (en) | 2023-03-28 |
Family
ID=83072638
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210619329.0A Active CN115015495B (en) | 2022-06-02 | 2022-06-02 | Dynamic close-range miniature intelligent sensor for quality of growing spherical fruits and vegetables |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115015495B (en) |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS63119612A (en) * | 1986-11-10 | 1988-05-24 | 株式会社クボタ | Guide apparatus of working machine for fruits and vegetables |
| JP2000199743A (en) * | 1998-05-15 | 2000-07-18 | Mitsui Mining & Smelting Co Ltd | Measuring apparatus for internal quality of object |
| JP2000292359A (en) * | 1999-04-02 | 2000-10-20 | Hisaichi Shibazaki | Internal quality inspection method of vegetable and fruit and its device |
| JP2002139433A (en) * | 2000-11-02 | 2002-05-17 | Kansai Tlo Kk | Method for discriminating internal quality of vegetable and fruit |
| JP2003098081A (en) * | 2001-09-27 | 2003-04-03 | Kubota Corp | Quality measuring device for agricultural products |
| JP2004180554A (en) * | 2002-12-02 | 2004-07-02 | National Agriculture & Bio-Oriented Research Organization | Method and apparatus for selectively harvesting fruit vegetables |
| JP2005009943A (en) * | 2003-06-18 | 2005-01-13 | Eminetto:Kk | Apparatus for measuring quality of vegetables and fruits |
| JP2007071779A (en) * | 2005-09-08 | 2007-03-22 | Mitsui Mining & Smelting Co Ltd | Rind-puffing identification device for produce, and rind-puffing identification method for produce |
| CN107094429A (en) * | 2017-07-03 | 2017-08-29 | 彭紫薇 | A kind of multiaxis fruit picking robot |
| CN107462528A (en) * | 2017-08-02 | 2017-12-12 | 塔里木大学 | A kind of method of multi-channel spectral detection jujube quality |
| CN111239684A (en) * | 2020-01-17 | 2020-06-05 | 中航华东光电（上海）有限公司 | Binocular fast distance measurement method based on YoloV3 deep learning |
| CN112781647A (en) * | 2021-01-07 | 2021-05-11 | 中国农业大学 | Miniature intelligent flight detector of fruit quality in growth |
| CN113063757A (en) * | 2021-02-02 | 2021-07-02 | 中国农业大学 | Quasi-spherical fruit and vegetable shape modeling method and optical map correction method |
| CN113902698A (en) * | 2021-09-30 | 2022-01-07 | 国网山西省电力公司 | Unmanned aerial vehicle holder progressive target focusing method based on intelligent visual control |
-
2022
- 2022-06-02 CN CN202210619329.0A patent/CN115015495B/en active Active
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS63119612A (en) * | 1986-11-10 | 1988-05-24 | 株式会社クボタ | Guide apparatus of working machine for fruits and vegetables |
| JP2000199743A (en) * | 1998-05-15 | 2000-07-18 | Mitsui Mining & Smelting Co Ltd | Measuring apparatus for internal quality of object |
| JP2000292359A (en) * | 1999-04-02 | 2000-10-20 | Hisaichi Shibazaki | Internal quality inspection method of vegetable and fruit and its device |
| JP2002139433A (en) * | 2000-11-02 | 2002-05-17 | Kansai Tlo Kk | Method for discriminating internal quality of vegetable and fruit |
| JP2003098081A (en) * | 2001-09-27 | 2003-04-03 | Kubota Corp | Quality measuring device for agricultural products |
| JP2004180554A (en) * | 2002-12-02 | 2004-07-02 | National Agriculture & Bio-Oriented Research Organization | Method and apparatus for selectively harvesting fruit vegetables |
| JP2005009943A (en) * | 2003-06-18 | 2005-01-13 | Eminetto:Kk | Apparatus for measuring quality of vegetables and fruits |
| JP2007071779A (en) * | 2005-09-08 | 2007-03-22 | Mitsui Mining & Smelting Co Ltd | Rind-puffing identification device for produce, and rind-puffing identification method for produce |
| CN107094429A (en) * | 2017-07-03 | 2017-08-29 | 彭紫薇 | A kind of multiaxis fruit picking robot |
| CN107462528A (en) * | 2017-08-02 | 2017-12-12 | 塔里木大学 | A kind of method of multi-channel spectral detection jujube quality |
| CN111239684A (en) * | 2020-01-17 | 2020-06-05 | 中航华东光电（上海）有限公司 | Binocular fast distance measurement method based on YoloV3 deep learning |
| CN112781647A (en) * | 2021-01-07 | 2021-05-11 | 中国农业大学 | Miniature intelligent flight detector of fruit quality in growth |
| CN113063757A (en) * | 2021-02-02 | 2021-07-02 | 中国农业大学 | Quasi-spherical fruit and vegetable shape modeling method and optical map correction method |
| CN113902698A (en) * | 2021-09-30 | 2022-01-07 | 国网山西省电力公司 | Unmanned aerial vehicle holder progressive target focusing method based on intelligent visual control |
Non-Patent Citations (2)
| Title |
|---|
| 彭彦昆 等: "樱桃番茄可溶性固形物含量的可见/近红外透射光谱无损检测", 《中国食品学报》 * |
| 彭彦昆 等: "苹果内部品质分级机械手设计与试验", 《农业机械学报第》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115015495B (en) | 2023-03-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111561873B (en) | Self-propelled seedling tree trunk phenotype information acquisition system and acquisition method thereof | |
| CN102384767B (en) | Nondestructive detection device and method for facility crop growth information | |
| CN202382768U (en) | Facility crop growth information nondestructive test apparatus | |
| US11436824B2 (en) | Water stress detection method for tomatoes in seedling stage based on micro-CT and polarization-hyperspectral imaging multi-feature fusion | |
| CN106406178B (en) | Real-time peer-to-peer monitoring device and monitoring method for greenhouse crop growth information | |
| CN101936882B (en) | Nondestructive testing method and device for nitrogen and water of crops | |
| CN103439265B (en) | Real-time monitoring method for growth characters of tea trees in intensive cultivation | |
| CN108858122A (en) | A kind of greenhouse plant disease crusing robot and method for inspecting | |
| CN103278503B (en) | Multi-sensor technology-based grape water stress diagnosis method and system therefor | |
| CN110065074A (en) | A kind of the visual servo laser orientation system and method for picking robot | |
| CN114037552B (en) | Method and system for polling physiological growth information of meat ducks | |
| CN104732564A (en) | Maize leaf area lossless dynamic monitoring device and method | |
| CN115015495B (en) | Dynamic close-range miniature intelligent sensor for quality of growing spherical fruits and vegetables | |
| CN116686545B (en) | Litchi picking robot shade removing method based on machine vision control | |
| CN112580513A (en) | Intelligent identification method for crop diseases and insect pests | |
| CN112816441A (en) | Method and device for detecting growth condition of facility horticultural crop | |
| Zhao et al. | Intelligent micro flight sensing system for detecting the internal and external quality of apples on the tree | |
| CN207622789U (en) | A kind of crop growth environment monitoring system based on sensor network | |
| CN105181632A (en) | Network-based multi-dimensional plant imaging NDVI measuring device | |
| Wang et al. | Research advance in phenotype detection robots for agriculture and forestry | |
| EP4280861A1 (en) | A data collection and monitoring system, a controlled environment farming system, devices and related methods | |
| CN115407771A (en) | Crop monitoring method, system and device based on machine vision | |
| CN114022907A (en) | Pig body surface temperature measuring device and method based on deep learning | |
| CN112781647A (en) | Miniature intelligent flight detector of fruit quality in growth | |
| KR102597253B1 (en) | Smart Farm System Through Video-based Crop Cultivation Growth Platform Construction |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |