CN113433131A - Physical multi-parameter coupled method and device for rapidly monitoring mildew of citrus peel - Google Patents
Physical multi-parameter coupled method and device for rapidly monitoring mildew of citrus peel Download PDFInfo
- Publication number
- CN113433131A CN113433131A CN202110793528.9A CN202110793528A CN113433131A CN 113433131 A CN113433131 A CN 113433131A CN 202110793528 A CN202110793528 A CN 202110793528A CN 113433131 A CN113433131 A CN 113433131A
- Authority
- CN
- China
- Prior art keywords
- mildew
- weight
- temperature
- citrus
- humidity
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
-
- 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
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N5/00—Analysing materials by weighing, e.g. weighing small particles separated from a gas or liquid
- G01N5/02—Analysing materials by weighing, e.g. weighing small particles separated from a gas or liquid by absorbing or adsorbing components of a material and determining change of weight of the adsorbent, e.g. determining moisture content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Mathematical Physics (AREA)
- Theoretical Computer Science (AREA)
- Pathology (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- Immunology (AREA)
- Computational Mathematics (AREA)
- Pure & Applied Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Algebra (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Food Science & Technology (AREA)
- Medicinal Chemistry (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention discloses a physical multi-parameter coupled method and a device for rapidly monitoring mildew of citrus peel, wherein the device comprises a closed box and a control layer; the two sides of the closed box are provided with vent holes, and a box body door of the closed box is provided with a glass observation door; the upper layer of the closed box is provided with a control layer; the inside of seal box is equipped with data acquisition system and data processing system, data acquisition system includes three subsystems: the image acquisition system, the weight acquisition system and the temperature and humidity acquisition system monitor the mildew condition of the citrus grandis samples and predict the quality condition of the citrus grandis by establishing an image identification model, a weight change mathematical model and a multi-parameter coupling model of the air temperature and humidity of the storage environment during use. The invention realizes the real-time monitoring and the timely early warning processing of the quality change of the citrus grandiflora in the warehousing process, provides technical support for safe and efficient warehousing, reduces the warehousing risk of the citrus grandiflora and reduces the warehousing cost.
Description
Technical Field
The invention belongs to the field of agricultural informatization, and particularly relates to a physical multi-parameter coupled method and device for rapidly monitoring mildew of citrus peel.
Background
The pericarpium citri reticulatae needs more than three years of aging process, and the value, the price and the aging storage time are in positive correlation, so the pericarpium citri reticulatae needs large-scale storage. Due to the medicinal and edible characteristics of the citrus grandiflora, the market of tea and health food related to the citrus grandiflora is rapidly expanded in recent years, so that the industrial scale of the citrus grandiflora is rapidly expanded, and the citrus grandiflora storage is required to be developed to scale and standardization. In the warehousing aging process of citrus grandis, the prevention of the citrus grandis from mildewing is one of the most important targets of the industrial control technology of the warehousing environment. Currently, acquiring citrus peel mildewing data is generally sampling inspection and mainly depends on biochemical means and related instrument and equipment for off-line detection, for example, patent CN106950298B discloses a method for simultaneously detecting mycotoxin and pesticide residue in citrus peel kumi at a new meeting, a modified multiwalled carbon nanotube is used as a purification adsorbent, and an improved QuEChERS method and a liquid chromatography-tandem mass spectrometer are combined and applied to detection of mycotoxin and pesticide in citrus peel kumi at the new meeting; but the preprocessing procedure is complex, the detection period is long, the detection result is delayed in time, and real-time feedback and closed-loop control on the storage environment control cannot be performed. Therefore, the real-time monitoring and early warning of the mildew of the pericarpium citri reticulatae is realized, so that the warehousing site of the pericarpium citri reticulatae is automatically controlled, and the real significance is very important.
Disclosure of Invention
The invention aims to overcome the defects in the prior art, provides an efficient and high-precision physical multi-parameter coupled citrus peel mildew rapid monitoring method and device, detects citrus peel mildew in real time based on a visual image, a citrus peel weight real-time acquisition and temperature and humidity acquisition system, and can be widely applied to citrus peel automated storage warehouses.
The purpose of the invention is realized by the following technical scheme:
a physical multi-parameter coupled citrus peel mildew rapid monitoring device is shown in figures 2 to 5 and comprises a closed box 1, a box body door 2, a glass observation door 3, a door hinge 4, a buzzer 5, a processor display panel 6, a temperature sensor 7, a humidity sensor 8, a vent hole 9, a real-time weight acquisition module 10, a weighing substrate 101, a pressure sensor 102, a weighing base 103, a weight data interface 104, a heating module 105, a camera 11, a light source 12, an image data interface 13, a processor 14 and a control layer 15.
The two sides of the closed box 1 are provided with vent holes 9 for simulating the actual warehouse environment and the ventilation conditions thereof; the box body door 2 is arranged on the closed box 1 through a door hinge 4, is opened when a citrus peel sample is placed, and is closed when monitoring is carried out, so that interference caused by external light source change is eliminated, and noise and errors of collected images are reduced; a glass observation door 3 is arranged at the central position of the box body door 2 and is used for manually observing the real-time state of the citrus peel sample inside; the upper layer of the closed box 1 is provided with a control layer 15; the inside of seal box 1 is equipped with data acquisition system and data processing system, data acquisition system includes three subsystems: the system comprises an image acquisition system, a weight acquisition system and a temperature and humidity acquisition system.
The data processing system comprises a processor 14, a buzzer 5 and a processor display panel 6, the processor 14 is installed inside a control layer 15, three data acquisition systems, namely an image acquisition system, a weight acquisition system and a temperature and humidity acquisition system, are communicated, target data to be sent are identified, extracted and subjected to decision analysis, the processor display panel 6 is fixedly installed on the outer side of the upper layer of a closed box 1 and used for displaying the temperature and the humidity in the box in real time, the buzzer 5 is embedded outside the control layer and connected with the processor 14 and used for sending buzzing for early warning after an instruction is obtained, further, the processor 14 is in remote communication with an upper computer for regulating and controlling a warehouse and making feedback, and the work and the operation of the warehouse dehumidifier are controlled.
The image acquisition system comprises a camera 11, a light source 12 and an image data interface 13, wherein the camera 11 is fixed at the top of the lower layer of the closed box 1 and used for acquiring the color real-time state images of the citrus peel at fixed intervals, the light source 12 is arranged beside the camera 11, and the light source 12 and the camera 11 are started synchronously and used for assisting the camera to acquire the images more clearly in a dark environment, so that errors caused by interference caused by reflection or ghost are avoided, the stability of the images is ensured, and the difficulty of image preprocessing is reduced; the image data interface 13 is used to communicate the processor 14 with the image acquisition system.
The weight acquisition system is a real-time weight acquisition module 10 and comprises a weighing base plate 101, a pressure sensor 102, a weighing base 103, a weight data interface 104 and a heating module 105; the pressure sensor 102 is arranged and nested in the center of the weighing base 103, the weighing base plate 101 and the pressure sensor 102 are coaxially arranged, the weight of the citrus peel placed on the base plate is converted into a digital signal, and the weight data interface 104 is arranged on the side surface of the weighing base 103 and is used for communicating the processor 14 to transmit data at regular time; for avoiding placing the container surface of wide tangerine peel sample under the environment of relative humidity, because the temperature difference changes a small amount of dew that condenses, causes the data error, set up heating module 105 in weighing base plate 101 bottom, be a little higher than ambient temperature in order to avoid weighing base plate 101 surface dewfall, guarantee the high accuracy and measure the weight, heating module 105 corresponds the heating region and installs, and heating module carries out the comparison through induction heating region temperature and ambient temperature, with the temperature control of heating region 0.2 ~ 0.5 ℃ higher than ambient temperature, heating module 105's control procedure flow chart is shown in figure 6.
The humiture collection system includes temperature sensor 7 and humidity transducer 8, all adopts the wall-hanging to install at the lower floor lateral wall of seal box 1.
A physical multi-parameter coupled citrus grandis mildew rapid monitoring method monitors mildew conditions of citrus grandis samples by establishing a citrus grandis image recognition model, a citrus grandis sample weight change mathematical model and a multi-parameter coupled model of storage environment temperature and humidity, and is used for predicting quality conditions of citrus grandis.
The physical multi-parameter coupled pericarpium citri reticulatae rapid mildew monitoring method is shown in a flow chart of fig. 1 and specifically comprises the following steps:
(1) storing the apparatus in a location consistent with the warehouse environmental conditions;
(2) placing the pericarpium citri reticulatae blanco sample on a weighing substrate 101, and starting a control program;
(3) triggering the camera 11 to capture a frame of static image at regular time, starting the LED light source during photographing, and turning off the LED light source at other time; the induction pressure sensor 102 records sample weight data at regular time; starting a temperature sensor 7 and a humidity sensor 8 to transmit environmental parameters in the box in real time;
(4) preprocessing image, weight and temperature and humidity data based on the collected data;
(5) extracting characteristic parameters of profile, color and texture data according to mould classification, establishing an image identification model, and performing classification identification;
(6) carrying out regression analysis by combining the weight data and the temperature and humidity parameters, and establishing a double-parameter fused weight change mathematical model;
(7) analyzing the weight mutation interval according to a first derivative equation of the fused weight fitting curve;
(8) performing information fusion on the processed physical parameter data according to a multi-criterion decision method, and establishing a multi-parameter coupling model;
(9) the processor 14 carries out classifier on-line judgment according to the multi-parameter coupling model, further triggers the buzzer to give an alarm, and the communication upper computer carries out feedback control on the warehouse to complete the rapid monitoring of the mildew of the integral citrus grandis.
In the step (2), according to the barrel effect, the orange peel of the sample Guangdong adopts the new red peel of the year, and the orange peel has high sugar content, is easy to mildew and has better early warning effect.
In the step (2), the heating module 105 comprises two temperature sensors for detection and is connected with the controller, one temperature sensor is used for detecting the temperature of the heating area, the other temperature sensor is used for detecting the ambient temperature and sending data to the controller, so that the temperature of the heating area is 0.2-0.5 ℃ higher than the ambient temperature.
In the step (3), the camera 11 is an industrial full-color camera, and target image data are collected once at an interval of 300-3600 s; the target weight data is collected once at intervals of 300-3600 s.
In the step (4), the preprocessing of the image data includes: image graying, image denoising, image segmentation and image enhancement.
In the step (5), the contour features are extracted by adopting a differential operator to obtain a sample citrus peel edge contour image, and a contour change degree value C1 is obtained.
In the step (5), color moment extraction is adopted for color features, color distribution in the citrus grandis peel image is obtained by calculating moment features, and the color change and distribution state C2 of the citrus grandis peel sample is obtained, wherein the formula is as follows:
in the formula, mui、σi、ξiRespectively a first moment, a second moment and a third moment of the image; pijIs the jth color component of the ith pixel; n is the number of pixels.
In the step (5), the texture features are extracted by adopting a gray level co-occurrence matrix T (NxN), and the contrast, the correlation and the entropy value in the defined step length and direction are calculated to obtain the texture feature value C3 of the surface of the pericarpium citri reticulatae.
In the step (5), the edge contour characteristic value C1, the color change and distribution state C2 of the sample pericarpium citri reticulatae and the texture characteristic value C3 of the surface of the pericarpium citri reticulatae are used as the index basis for image judgment to establish an image recognition model, the three characteristic values are input into an SVM classifier in the processor 14 for classification and recognition, and the existing mildew characteristic data are compared to obtain a classification result:
in the step (6), the weight change of the pericarpium citri reticulatae of the sample is influenced by the coupling of mildew and temperature and humidity factors, so as to prevent the sudden change of the temperature and the humidity of the air from causingThe sample quality mutation is classified into mildew by the system, the influence weight of sample data of weight and temperature and humidity in the whole data needs to be determined, and the change equation F of the weight of the citrus peel is obtained according to the actual sample1(x) Equation F of dynamic response of normal pericarpium Citri Reticulatae Chachiensis weight varying with temperature and humidity2(x) And obtaining a two-parameter fused weight change mathematical model: g (x) ═ F1(x)-aF2(x) (ii) a Wherein a is a temperature and humidity parameter weight coefficient, and the value range of a is 0.20-0.45.
In the step (7), the first derivative equation of the fused weight fitting curve is as follows:
the first derivative equation can reflect the data change speed, so that the mutation interval of the weight data is determined, and the sample mildew characteristics are extracted.
In the step (8), the image recognition model and the weight change mathematical model are used as two independent mildew characterization monitoring methods, and the obtained multi-parameter coupling model is as follows:
in the step (9), inputting and identifying the fused information model features through an SVM classifier:
wherein
For AND gate logic, a 1 is output only if the mildew phenomena are satisfied simultaneously with C (x) and F (x).
The principle of the invention is as follows: because the mold grows rapidly during the initial mildewing period of the citrus chachiensis, the thalli absorb external substances such as air moisture and the like to increase the quality of the citrus chachiensis, the dynamic quality monitoring data of the citrus chachiensis monitoring sample has mutation characteristics and is different from the normal quality change of the citrus chachiensis sample caused by the change of the environmental temperature and humidity; and the velvet hyphae are formed on the surface of the citrus peel at the initial stage of the generation of the mould, and have obvious characteristics visually; therefore, the method adopts physical multi-parameter coupling to rapidly monitor the mildew of the citrus grandis, and can realize real-time monitoring and timely early warning processing of the quality change of the citrus grandis in the warehousing process.
Compared with the prior art, the invention has the following advantages and effects:
(1) the physical multi-parameter coupled pericarpium citri reticulatae rapid mildew monitoring device disclosed by the invention realizes real-time monitoring and timely early warning processing of pericarpium citri reticulatae quality change in the warehousing process, provides technical support for safe and efficient warehousing, reduces warehousing risks of pericarpium citri reticulatae and reduces warehousing cost.
(2) According to the physical multi-parameter coupled citrus peel mildew rapid monitoring device, a machine vision technology is utilized to capture a static image of a citrus peel sample for processing, and the mildew degree is judged according to the characteristics of the contour, the color or the texture; processing physical parameters of the citrus peel samples to judge the mildew degree by utilizing the coupling of weight and temperature and humidity parameters; the citrus peel mildew characteristic parameters are collected from different aspects to make decisions, and the precision is high.
(3) According to the physical multi-parameter coupled rapid citrus peel mildew monitoring device, the intelligent heating module is additionally arranged on the weight acquisition system, so that the measurement error caused by condensation on the scale pan due to temperature difference in a humid environment is avoided, and high-precision acquisition is realized.
(4) According to the physical multi-parameter coupled pericarpium citri reticulatae mouldness rapid monitoring device, the exclusive storage technical specification of the storage, aging and quality guarantee requirements of pericarpium citri reticulatae can be established at the later stage according to the integrated analysis of alarm critical data, and an intelligent pericarpium citri reticulatae warehouse is built by modern scientific technology.
Drawings
FIG. 1 is a flow chart of a physical multi-parameter coupled method for rapidly monitoring citrus peel mildew.
Fig. 2 is a schematic structural diagram of a physical multi-parameter coupled rapid citrus peel mildew monitoring device.
Fig. 3 is a schematic half-section of the apparatus.
Fig. 4 is a schematic structural diagram of a real-time weight acquisition module.
FIG. 5 is a schematic view of a heating module.
FIG. 6 is a flow chart illustrating the operation of the heating module.
FIG. 7-a is a graph showing the change in weight of normal pericarpium Citri Reticulatae Chachiensis (the new skin of the same year) when its humidity is decreasing with time while its temperature is relatively stable.
FIG. 7-b is a graph showing the change in weight of normal pericarpium Citri Reticulatae Chachiensis (the new skin of the same year) when its humidity is increasing with time while its temperature is stable.
Wherein, 1, closing the box; 2. a box door; 3. a glass observation door; 4. a door hinge; 5. a buzzer; 6. a processor display panel; 7. a temperature sensor; 8. a humidity sensor; 9. a vent hole; 10. a real-time weight acquisition module; 101. weighing the substrate; 102. a pressure sensor; 103. a weighing base; 104. a weight data interface; 105. a heating module; 11. a camera; 12. a light source; 13. an image data interface; 14. a processor; 15. and controlling the layer.
Detailed Description
In order that the invention may be readily understood, reference will now be made in detail to the specific embodiments of the invention. The following examples will assist those skilled in the art in further understanding the invention, but are not intended to limit the invention in any way. It should be noted that, for a person skilled in the art, many variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention.
Example 1
A physical multi-parameter coupled citrus chachiensis ashmead rapid monitoring device is shown in figures 2, 3, 4 and 5, wherein ventilation holes 9 are formed in two sides of a closed box 1, a box body door 2 is installed on the closed box 1 through a door hinge 4, a glass observation door 3 is arranged in the center of the box body door 2, and a control layer 15 is arranged on the upper layer of the closed box 1; the inside of seal box 1 is equipped with data acquisition system and data processing system, data acquisition system includes three subsystems: the system comprises an image acquisition system, a weight acquisition system and a temperature and humidity acquisition system. The humiture collection system includes temperature sensor 7 and humidity transducer 8, all adopts the wall-hanging to install at the lower floor lateral wall of seal box 1. The data processing system comprises a processor 14, a buzzer 5 and a processor display panel 6, wherein the processor 14 is installed inside a control layer 15, three data acquisition systems, namely an image acquisition system, a weight acquisition system and a temperature and humidity acquisition system, are communicated, the processor display panel 6 is fixedly installed on the outer side of the upper layer of the closed box 1 and used for displaying the temperature and the humidity in the box in real time, and the buzzer 5 is embedded outside the control layer and connected with the processor 14 and used for sending buzzing for early warning after an instruction is obtained. The image acquisition system comprises a camera 11, a light source 12 and an image data interface 13, wherein the camera 11 is fixed at the top of the lower layer of the closed box 1 and is positioned right above the weighing substrate 101, full-color images under the real-time state of the citrus reticulata blanco are automatically shot at intervals of 15min, the LED lamp group light source 12 is arranged beside the camera 11, and the image data interface 13 is used for communicating the processor 14 with the image acquisition system. The weight acquisition system is a real-time weight acquisition module 10 and comprises a weighing base plate 101, a pressure sensor 102, a weighing base 103, a weight data interface 104 and a heating module 105; the pressure sensor 102 is installed and nested in the center of the weighing base 103, the pressure sensor 102 converts weight into digital signals, weight data of the citrus peel in a real-time state are automatically recorded every 15min, the weighing base plate 101 and the pressure sensor 102 are installed coaxially, a weight data interface 104 is installed on the side face of the weighing base 103, a heating module 105 is arranged at the bottom of the weighing base plate 101, and the temperature of a heating area is controlled to be 0.2-0.5 ℃ higher than the ambient temperature by sensing the temperature of the heating area and comparing the temperature with the ambient temperature.
When the citrus peel storage mildew early warning is carried out, as shown in fig. 1, a citrus peel sample is placed on a weighing substrate 101, a heating module 105 is triggered, two temperature measuring sensors respectively sense the temperature of a heating area and the ambient temperature and compare the temperatures, the temperature of an electric heating wire is regulated and controlled, and the temperature of the heating area is controlled to be 0.2-0.5 ℃ higher than the ambient temperature so as to ensure the weighing precision, as shown in fig. 6; the camera 11 starts an LED light source when working, captures a frame of static image at regular time, transmits the static image to the processor 14, preprocesses the image, namely graying, filtering and binaryzation, extracts characteristic parameters of contour, color and texture of the citrus peel of the sample, wherein the contour extraction adopts a differential operator to compare perimeter change, a color moment to compare color distribution and a gray level co-occurrence matrix to compare and represent texture states, and establishes an image recognition model according to three visual characteristic values. Because the increase of the weight of the citrus grandis is influenced by the coupling of mildew and temperature and humidity parameters, the collected weight change data and the citrus grandis dynamic water content data are combined and subjected to first-order derivation, a weight mutation interval is found, and whether the change is caused by mildew or not is judged. Starting a pressure sensor 102 to record weight data at regular time, simultaneously transmitting environmental parameters in the box by a temperature sensor 7 and a humidity sensor 8 in real time, setting the weight proportion of the influence of the environmental parameters on the weight by a processor 14, and analyzing specific reasons of weight mutation intervals to obtain a weight change mathematical model; and (3) passing the multi-parameter coupling model through a classifier of AND gate logic, triggering a buzzer to alarm when the parameters simultaneously meet the mildew phenomenon, and finishing the operation of quickly monitoring the mildew of the citrus peel.
Example 2
A physical multi-parameter coupled citrus peel mildew rapid monitoring device can be used for citrus peel mildew early warning. When in use, the medicine is as follows: the characteristic value model C (x) of the citrus peel of the sample at a certain stage meets the mildew characteristic, when aspergillus flavus appears on the surface of the citrus peel, the citrus peel may grow on the edge contour to cause the side length of an image to be enlarged, the color of the citrus peel obviously changes to be green, and the characteristic texture is villous. And (3) judging a weight change mathematical model, wherein the weight change of the citrus peel is influenced by mildew and temperature and humidity factors in a coupling manner, and setting a temperature and humidity weight coefficient to be 0.20-0.45 for preventing mutation caused by air humidity from being classified into mildew by a system by mistake. As shown in fig. 7, the dynamic response curve of the citrus peel under normal conditions with weight varying with temperature and humidity shows that there is no mutation area, which indicates that the citrus peel does not mildew at this stage, and no early warning is needed; if the weight change of the citrus peel is 0.313g in the measurement stage, subtracting the temperature and humidity caused part to obtain 0.1522g of the actual weight mutation interval, obtaining F (x) meeting the mildew phenomenon, triggering a buzzer to alarm, and completing the rapid monitoring operation of the citrus peel mildew.
The above description is only an example of the present invention, but the present invention is not limited to the above example, and any other changes, modifications, substitutions, combinations, and simplifications which do not depart from the spirit and principle of the present invention and are equivalent to each other are included in the protection scope of the present invention.
Claims (10)
1. The utility model provides a wide tangerine peel of physics multi-parameter coupling mildenes and rot quick monitoring devices which characterized in that: comprises a closed box and a control layer; the two sides of the closed box are provided with vent holes, and a box body door of the closed box is provided with a glass observation door; the upper layer of the closed box is provided with a control layer; the inside of seal box is equipped with data acquisition system and data processing system, data acquisition system includes three subsystems: the system comprises an image acquisition system, a weight acquisition system and a temperature and humidity acquisition system.
2. The citrus peel mildew rapid monitoring device according to claim 1, characterized in that: the data processing system comprises a processor, a buzzer and a processor display panel; the processor is arranged in the control layer and is communicated with three data acquisition systems, namely an image acquisition system, a weight acquisition system and a temperature and humidity acquisition system; the processor display panel is fixedly arranged on the outer side of the upper layer of the closed box; the buzzer is embedded outside the control layer and connected with the processor.
3. The citrus peel mildew rapid monitoring device according to claim 1, characterized in that: the temperature and humidity acquisition system comprises a temperature sensor and a humidity sensor which are mounted on the side wall of the lower layer of the closed box in a wall-hanging manner; the image acquisition system comprises a camera, a light source and an image data interface, wherein the camera is fixed at the top of the lower layer of the closed box, the light source is arranged beside the camera, and the image data interface is used for communicating the processor with the image acquisition system.
4. The citrus peel mildew rapid monitoring device according to claim 1, characterized in that: the weight acquisition system is a real-time weight acquisition module and comprises a weighing substrate, a pressure sensor, a weighing base, a weight data interface and a heating module; the pressure sensor is arranged and nested in the center of the weighing base, the weighing substrate and the pressure sensor are coaxially arranged, and the weight data interface is arranged on the side surface of the weighing base; the heating module is arranged at the bottom of the weighing substrate, comprises two temperature measuring sensors and is connected with the controller, one temperature measuring sensor is used for detecting the temperature of the heating area, the other temperature measuring sensor is used for detecting the ambient temperature, the heating module controls the temperature of the heating area to be 0.2-0.5 ℃ higher than the ambient temperature by sensing the temperature of the heating area and comparing the temperature with the ambient temperature.
5. A physical multi-parameter coupled method for rapidly monitoring mildew of citrus peel is characterized by comprising the following steps: the method is characterized in that a citrus grandiflora image recognition model, a citrus grandiflora sample weight change mathematical model and a multi-parameter coupling model of storage environment temperature and humidity are established, so that the mildew condition of citrus grandiflora samples is monitored, and the method is used for predicting the quality condition of citrus grandiflora.
6. The method for rapidly monitoring the mildew of citrus chachiensis according to claim 5, wherein: the citrus peel mildew rapid monitoring device is adopted, and comprises the following steps:
(1) storing the apparatus in a location consistent with the warehouse environmental conditions;
(2) placing the pericarpium citri reticulatae blanco sample on a weighing substrate, and starting a control program;
(3) triggering a camera to capture a frame of static image at regular time, and starting a light source during photographing; the pressure sensor records sample weight data at regular time; starting a temperature sensor and a humidity sensor to transmit environmental parameters in the box in real time;
(4) preprocessing image, weight and temperature and humidity data based on the collected data;
(5) extracting characteristic parameters of profile, color and texture data according to mould classification, establishing an image identification model, and performing classification identification;
(6) carrying out regression analysis by combining the weight data and the temperature and humidity parameters, and establishing a double-parameter fused weight change mathematical model;
(7) analyzing the weight mutation interval according to a first derivative equation of the fused weight fitting curve;
(8) performing information fusion on the processed physical parameter data according to a multi-criterion decision method, and establishing a multi-parameter coupling model;
(9) the processor performs classifier on-line judgment according to the multi-parameter coupling model, further triggers the buzzer to give an alarm, and the communication upper computer performs feedback control on the warehouse to complete the rapid monitoring of the mildew of the integral citrus peel.
7. The method for rapidly monitoring the mildew of citrus chachiensis according to claim 6, wherein: in the step (5), extracting the contour features by adopting a differential operator to obtain a sample citrus peel edge contour image and obtain a contour change degree value C1; color characteristics are extracted by adopting color moments, color distribution in the citrus peel image is obtained by calculating the moment characteristics, and the color change and distribution state C2 of the citrus peel is obtained, wherein the formula is as follows:
in the formula, mui、σi、ξiRespectively a first moment, a second moment and a third moment of the image; pijIs the jth color component of the ith pixel; n is the number of pixels;
extracting texture features by adopting a gray level co-occurrence matrix T (NXN), calculating contrast, correlation and entropy values in the defined step length and direction, and obtaining a texture feature value C3 of the surface of the pericarpium citri reticulatae; then, establishing an image recognition model by taking the edge contour characteristic value C1, the color change and distribution state C2 of the sample pericarpium citri reticulatae and the texture characteristic value C3 of the surface of the pericarpium citri reticulatae as index bases for image judgment, inputting the three characteristic values into an SVM classifier in a processor for classification and recognition, and comparing the existing mildew characteristic data to obtain a classification result:
8. the method for rapidly monitoring the mildew of citrus chachiensis according to claim 6, wherein: in the step (6), the weight change of the pericarpium citri reticulatae is influenced by the coupling of mildew and temperature and humidity factors, in order to prevent sample quality mutation caused by sudden change of air temperature and humidity from being classified into mildew by a system by mistake, the influence weight of sample data of weight and temperature and humidity in the whole data needs to be determined, and according to the change equation F of the actual weight of the pericarpium citri reticulatae, the pericarpium citri reticulatae is subjected to1(x) Equation F of dynamic response of normal pericarpium Citri Reticulatae Chachiensis weight varying with temperature and humidity2(x) And obtaining a two-parameter fused weight change mathematical model: g (x) ═ F1(x)-aF2(x) (ii) a Wherein a is a temperature and humidity parameter weight coefficient, and the value range of a is 0.20-0.45; in the step (7), the first derivative equation of the fused weight fitting curve is as follows:
therefore, the mutation interval of the weight data is determined, and the mildew characteristics of the sample are extracted.
9. The method for rapidly monitoring the mildew of citrus chachiensis according to claim 6, wherein: in the step (8), the image recognition model and the weight change mathematical model are used as two independent mildew characterization monitoring methods, and the obtained multi-parameter coupling model is as follows:
10. the method for rapidly monitoring the mildew of citrus chachiensis according to claim 6, wherein: in the step (9), inputting and identifying the fused information model features through an SVM classifier:
wherein
For AND gate logic, a 1 is output only if the mildew phenomena are satisfied simultaneously with C (x) and F (x).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110793528.9A CN113433131A (en) | 2021-07-14 | 2021-07-14 | Physical multi-parameter coupled method and device for rapidly monitoring mildew of citrus peel |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110793528.9A CN113433131A (en) | 2021-07-14 | 2021-07-14 | Physical multi-parameter coupled method and device for rapidly monitoring mildew of citrus peel |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113433131A true CN113433131A (en) | 2021-09-24 |
Family
ID=77760334
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110793528.9A Pending CN113433131A (en) | 2021-07-14 | 2021-07-14 | Physical multi-parameter coupled method and device for rapidly monitoring mildew of citrus peel |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113433131A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114486603A (en) * | 2022-01-14 | 2022-05-13 | 华南农业大学 | Method and equipment for detecting dynamic water content of citrus grandis |
| CN116482311A (en) * | 2023-04-28 | 2023-07-25 | 江门丽宫国际食品股份有限公司 | Aging method and device for dynamic quality monitoring of critical state of pericarpium citri reticulatae |
| CN117142687A (en) * | 2023-08-29 | 2023-12-01 | 江苏国强环保集团有限公司 | Intelligent online monitoring system and intelligent online monitoring process for desulfurization wastewater |
-
2021
- 2021-07-14 CN CN202110793528.9A patent/CN113433131A/en active Pending
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114486603A (en) * | 2022-01-14 | 2022-05-13 | 华南农业大学 | Method and equipment for detecting dynamic water content of citrus grandis |
| CN114486603B (en) * | 2022-01-14 | 2023-08-29 | 华南农业大学 | Method and equipment for detecting dynamic water content of pericarpium citri reticulatae |
| CN116482311A (en) * | 2023-04-28 | 2023-07-25 | 江门丽宫国际食品股份有限公司 | Aging method and device for dynamic quality monitoring of critical state of pericarpium citri reticulatae |
| CN116482311B (en) * | 2023-04-28 | 2023-11-24 | 江门丽宫国际食品股份有限公司 | Aging method and device for dynamic quality monitoring of critical state of pericarpium citri reticulatae |
| CN117142687A (en) * | 2023-08-29 | 2023-12-01 | 江苏国强环保集团有限公司 | Intelligent online monitoring system and intelligent online monitoring process for desulfurization wastewater |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113433131A (en) | Physical multi-parameter coupled method and device for rapidly monitoring mildew of citrus peel | |
| CN105300854B (en) | Droplet parameter measuring apparatus and the droplet parameter measurement analysis method for utilizing the device | |
| CN110133202A (en) | A kind of method and device of the food materials monitoring freshness of intelligent kitchen | |
| CN102589705B (en) | Single-blackbody temperature-controlled MRTD (Minimum Resolvable Temperature Difference) field online automatic detection device and method for thermal imager | |
| CN215985764U (en) | Physical multi-parameter coupled quick monitoring device for mildew of citrus peel | |
| Pan et al. | Eggshell crack detection based on computer vision and acoustic response by means of back-propagation artificial neural network | |
| Zhu et al. | Evaluation of green tea sensory quality via process characteristics and image information | |
| CN106529442A (en) | Pedestrian identification method and apparatus | |
| CN108172288A (en) | Medical Devices intelligent control method, device and computer readable storage medium | |
| CN112669305B (en) | Metal surface rust resistance test bench and rust resistance evaluation method | |
| CN109307675A (en) | A kind of product appearance detection method and system | |
| CN111189494A (en) | Measurement label, image color restoration method thereof and measurement identification method | |
| CN112415050B (en) | Mutton mixing supposition discriminating method based on temperature distribution difference | |
| CN201035376Y (en) | Failure diagnosis device under small sample conditional in the process of manufacturing production | |
| CN109842682A (en) | A kind of study of distributed environment safety and method for early warning based on Internet of Things | |
| CN116778408A (en) | Intelligent monitoring and early warning platform for safety production of chemical industry park based on digital twinning | |
| CN114970675A (en) | Artificial nose refrigerator food freshness detection system and method based on feature selection | |
| WO2019224707A1 (en) | A method for the remote detection of the internal temperature of a cabin of an aircraft through at least one thermal imaging camera, and means for carrying out such method | |
| Martynenko et al. | An intelligent control system for thermal processing of biomaterials | |
| CN115378968A (en) | Biological laboratory safety detection system based on edge calculation | |
| CN114880361A (en) | Production, processing, supervision and management method and system for bladder fistulation tube | |
| CN111061206B (en) | Quality inspection operation remote monitoring early warning system | |
| CN111398334A (en) | Novel melting point sliding point instrument and correction measurement method thereof | |
| CN114219750A (en) | Method for quantitatively detecting meat quality based on fusion of thermal imaging and visual imaging in time sequence temperature changing process | |
| CN117891159A (en) | Food processing control system and method based on big data |
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 |