CN112560575A - Red Fuji apple shape data enhancement device and method - Google Patents
Red Fuji apple shape data enhancement device and method Download PDFInfo
- Publication number
- CN112560575A CN112560575A CN202011236498.3A CN202011236498A CN112560575A CN 112560575 A CN112560575 A CN 112560575A CN 202011236498 A CN202011236498 A CN 202011236498A CN 112560575 A CN112560575 A CN 112560575A
- Authority
- CN
- China
- Prior art keywords
- apple
- apples
- data set
- data
- pictures
- 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
- 238000000034 method Methods 0.000 title claims abstract description 26
- 241000220225 Malus Species 0.000 claims abstract description 279
- 235000021016 apples Nutrition 0.000 claims abstract description 92
- 235000010724 Wisteria floribunda Nutrition 0.000 claims abstract description 32
- 238000012549 training Methods 0.000 claims abstract description 26
- 230000002708 enhancing effect Effects 0.000 claims abstract description 21
- 230000005540 biological transmission Effects 0.000 claims description 31
- 230000007246 mechanism Effects 0.000 claims description 30
- 238000012545 processing Methods 0.000 claims description 24
- 235000013399 edible fruits Nutrition 0.000 claims description 18
- 230000006870 function Effects 0.000 claims description 14
- 238000012544 monitoring process Methods 0.000 claims description 13
- 230000009467 reduction Effects 0.000 claims description 12
- 230000004888 barrier function Effects 0.000 claims description 11
- 239000013589 supplement Substances 0.000 claims description 9
- 238000001914 filtration Methods 0.000 claims description 8
- 230000008569 process Effects 0.000 claims description 6
- 238000007664 blowing Methods 0.000 claims description 5
- 239000000284 extract Substances 0.000 claims description 4
- 239000006185 dispersion Substances 0.000 claims description 3
- 239000000126 substance Substances 0.000 claims description 3
- 230000003014 reinforcing effect Effects 0.000 claims 1
- 238000010586 diagram Methods 0.000 description 5
- 238000011160 research Methods 0.000 description 5
- 238000007405 data analysis Methods 0.000 description 3
- 238000013461 design Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000005070 sampling Methods 0.000 description 3
- 238000013473 artificial intelligence Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 241000282414 Homo sapiens Species 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000003466 anti-cipated effect Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000003416 augmentation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 235000012055 fruits and vegetables Nutrition 0.000 description 1
- JEIPFZHSYJVQDO-UHFFFAOYSA-N iron(III) oxide Inorganic materials O=[Fe]O[Fe]=O JEIPFZHSYJVQDO-UHFFFAOYSA-N 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001502 supplementing effect Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/70—Circuitry for compensating brightness variation in the scene
- H04N23/74—Circuitry for compensating brightness variation in the scene by influencing the scene brightness using illuminating means
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N5/00—Details of television systems
- H04N5/76—Television signal recording
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
- G06T2207/20032—Median filtering
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Signal Processing (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
- Sorting Of Articles (AREA)
Abstract
The embodiment of the invention provides a device and a method for enhancing appearance data of red Fuji apples, wherein under the control of a controller on the following devices, an apple is photographed by an appearance data acquisition device to form an original data set; classifying and calibrating the pictures in the original data set by a data classification device to obtain a calibration data set; dividing an interested area of a picture in a calibration data set by a data cutting device, cutting and eliminating noise and reserving the interested area to form a sample data set; and inputting the sample data set into a deep convolution generation countermeasure network through a picture generation device, training to generate a new apple picture, and forming a result data set to obtain enhanced data of the appearance of the Fuji apple.
Description
Technical Field
The invention relates to the technical field of artificial intelligence and deep learning, in particular to a device and a method for enhancing appearance data of red Fuji apples.
Background
The apple industry is one of the important fruit and vegetable industries in China, and the research on the appearance of the apples is an important link for improving the quality of the apples and promoting the forward development of the apple industry. But at present, large-scale apple picture data with good quality and complete grading is lacked when relevant research is carried out on apples. The apple appearance data that most research institute adopted are mostly the little sample data that the researcher shot by oneself, and the means of shooting mostly adopts artifical manual shooting, and this kind of mode work efficiency is low, work load is big, the shooting environment is not standard to often can cause the appearance data quality who obtains to differ because factors such as photographer's people are artificial, and appear easily that the factor data sample volume undersize makes whole research can't reach the anticipated purpose, seriously influences subsequent data analysis.
Disclosure of Invention
The embodiment of the invention provides a device and a method for enhancing the appearance data of red Fuji apples, which realize the automatic acquisition of the appearance data of the red Fuji apples by automatically photographing and acquiring photos and automatically analyzing and processing the photo data and obtain the amplified photo data by an artificial intelligence method, thereby providing more sufficient data samples for the subsequent data analysis.
In order to achieve the above object, in one aspect, an embodiment of the present invention provides a red fuji apple shape data enhancing apparatus, including:
the appearance data acquisition device 101 is used for photographing the apples and forming an original data set by the photographed pictures;
the data classification device 102 is configured to classify and calibrate the pictures in the original data set to obtain a calibrated data set;
the data cutting device 103 is used for defining an interested region for the pictures in the calibration data set, cutting and reserving the interested region, performing median filtering on the interested region, and forming a sample data set by the pictures stored in the interested region after noise is eliminated;
the picture generation device 104 is used for inputting the sample data set into a deep convolution generation countermeasure network, performing training processing, generating a new apple picture and forming a result data set;
a controller 105 for controlling the functions of the shape data acquiring device 101, the data sorting device 102, the data cropping device 103 and the picture generating device 104, and storing the data in the execution process of the devices in the memory of the controller.
Further, the shape data acquisition device 101 includes: the device comprises a frame 1, an apple transmission device, an apple feeding mechanism, an appearance collecting device and an apple position adjusting device;
the apple transmission device is arranged on the upper surface of the frame 1 and extends along the upper surface of the frame 1;
the apple feeding mechanism is arranged at one end of the apple transmission device and stretches across the upper part of the apple transmission device;
the appearance acquisition device is arranged at the other end of the apple transmission device and stretches across the upper part of the apple transmission device;
the apple position adjusting device is arranged on the apple transmission device and is positioned below the appearance acquisition device.
Further, the apple gear includes: a conveyor belt 21, a barrier rib 22;
the conveyor belt 21 is positioned on the upper surface of the rack 1, and the conveying direction of the conveyor belt 21 is that the conveyor belt translates from the apple feeding mechanism to the shape acquisition device along the upper surface of the rack 1;
the stop bars 22 are vertically arranged on the upper surface of the frame 1 and are close to two side edges of the conveyor belt 21; the distance between the two barrier strips 22 is greater than the maximum diameter of the apple;
wherein the conveyor belt 21 includes: a recess 211;
the grooves 211 are used for containing apples, are hemispherical, and are uniformly arranged on the conveyor belt 21 along the conveying direction of the conveyor belt 21;
the groove 211 includes: a vent 2111;
the vent 2111 is disposed within the recess 211.
Further, apple feeding mechanism includes: a cylindrical hopper 31;
the cylindrical feed hopper 31 includes: a stepping motor 311, a rotating shaft 312, a dispersing air bag 313, a rotating impeller 314, and a through hole 315;
wherein the content of the first and second substances,
the top of the cylindrical feed hopper 31 is provided with a feed inlet;
the stepping motor 311 is installed at the upper side inside the cylindrical feed hopper 31;
one end of the rotating shaft 312 is connected to the stepping motor 311, and the other end of the rotating shaft 312 extends toward the bottom of the cylindrical hopper 31;
the dispersion air bag 313 is connected to one end of the rotating shaft 312 near the bottom of the cylindrical feed hopper 31; the dispersing air bag 313 is used for preliminarily dispersing the apples poured from the feeding hole into the rotating impeller 314;
the number of the rotating impellers 314 is three, and the three rotating impellers 314 are uniformly distributed and fixed at one end of the rotating shaft 312, which is close to the bottom of the cylindrical feed hopper 31, and are positioned below the dispersing air bags 313; an apple outlet is formed in the side wall of the rotary impeller 314;
the through-hole 315 is located the bottom of cylindricality feeder hopper 31 is close to on the lateral wall of appearance collection system one side, just the center of through-hole 315 with on the conveyer belt 21 the line of central connecting of recess 211 aligns, and works as rotatory impeller 314 rotates to just right when the through-hole 315, the center of through-hole 315 with the center of apple export aligns.
Further, the shape acquisition device includes: a gate frame 41;
the door-shaped frame 41 spans the stop strip 22;
the door-shaped frame 41 is provided with a camera 411, a motor 412, a fill-in light 413 and a monitoring video recorder 414;
the camera 411 is used for taking a picture of the apple;
an output shaft of the motor 412 is connected with a telescopic rod; the telescopic rod is connected with a black rubber plate; the telescopic rod is used for adjusting the position of the black rubber plate under the driving of the motor 412; the black rubber plate is used for providing a black background and intercepting the apples to be photographed subsequently on the conveyor belt 21 when the camera 411 takes photos for the apples;
the fill light 413 is used for providing ambient brightness when the camera 411 shoots;
the monitoring recorder 414 is used for recording the apple to be photographed to obtain the position of the apple to be photographed exposed to the shooting range of the camera 411.
Further, the apple position adjusting device includes: a clamping mechanism and an air tap;
the clamping mechanism is positioned below the appearance acquisition device;
the fixture includes: two clamping baffles 5111, four shifting inclined plates 5112 and two electric push rods 512;
the two clamping baffles 5111 are symmetrically arranged on two sides of the central connecting line of the groove 211 on the conveyor belt 21 in parallel and are positioned on the inner sides of the two stop strips 22;
the four shifting inclined plates 5112 are respectively connected to two ends of the two clamping baffles 5111, one end of any shifting inclined plate 5112 is connected to the clamping baffle 5111, and the other end of the shifting inclined plate 5112 extends in a direction obliquely adjacent to the stop strip 22;
one end of each electric push rod 512 is connected with one clamping baffle 5111, and the other end of each electric push rod 512 penetrates through the stop strip 22 adjacent to the clamping baffle 5111 and is installed on the rack 1 through a connecting seat;
the air nozzle is positioned below the conveyor belt 21 and is aligned with a central connecting line of the groove 211 on the conveyor belt 21; the air tap is connected with the air pump; an electromagnetic control valve is arranged on a connecting pipeline of the air tap and the air pump; the solenoid control valve is controlled by the controller; the air tap is used for blowing air to the apple through the vent 2111 to finely adjust the position of the apple.
Further, the data classification device specifically includes:
the processing unit of the controller identifies and extracts the texture and shape information of the pictures in the original data set, divides the pictures in the original data set into three types of special fruits, first fruits and second fruits, and respectively calibrates the divided three types of apple pictures.
Further, the picture generation apparatus includes: a generator and a discriminator;
the generator is used for performing upsampling by adopting fractional step convolution;
the discriminator is used for performing downsampling by adopting integer step length convolution;
the picture generation device is specifically configured to:
selecting a type of pictures subjected to noise reduction processing and inputting the pictures into a generator, wherein the generator generates a type of new pictures; simultaneously inputting the pictures subjected to noise reduction processing and the generated new pictures into the discriminator; fixing the generator, training the discriminator, stopping training the discriminator when the output value of the corresponding loss function reaches a preset convergence value, further fixing the discriminator, training the generator, stopping training the generator when the output value of the corresponding loss function reaches the preset convergence value, and repeating the steps until the generator and the discriminator model reach a Nash equilibrium state by continuously updating the parameters of the generator and the discriminator.
On the other hand, the embodiment of the invention adopts a red Fuji apple shape data enhancement device, and provides a red Fuji apple shape data enhancement method, which comprises the following steps:
and 204, inputting the sample data set into a deep convolution generation countermeasure network, performing training processing, and generating a new apple picture to form a result data set.
Further, the apple is photographed, and the photographed pictures form an original data set, specifically:
the apples are poured into the feeding hole, the controller 105 controls the stepping motor 311 to drive the rotary impeller 314 to rotate slowly, the apples are dispersed by the dispersing air bags 313 and enter the rotary impeller 314, the apples enter the conveyor belt 21 through the apple outlets formed in the side wall of the rotary impeller 314 and the through holes 315 in the cylindrical feed hopper 31 one by one, when the apples in the groove 211 in the conveyor belt 21 pass through the lower end of the portal frame 41, the controller 105 controls the output shaft of the motor 412 to drive the telescopic rod to extend, the telescopic rod drives the black rubber plate to move downwards, and meanwhile, the electric push rod 512 adjusts the telescopic length of the electric push rod 512 according to the diameter of the apples conveyed to the lower end of the portal frame 41, so that the position of the apples is limited; the controller 105 controls the monitoring video recorder 414 to record the position of the apple, the acquired data is uploaded to the controller 105, the controller 105 compares the acquired data with preset data in the controller 105 to determine whether the position of the apple exposed in the shooting range of the camera 411 is a preset area, if the position of the apple exposed in the shooting range of the camera 411 in the groove 211 is not in the preset area, the controller 105 gives the size of the apple deviating from the preset area, meanwhile, the controller 105 controls an air nozzle at the bottom of the corresponding conveyor belt 21 to work, fine adjustment is carried out on the position of the apple until the shooting area of the apple is matched with the preset area, and the air nozzle is closed; meanwhile, the controller 105 adjusts the brightness of the light supplementing lamp 413 to supplement light for the apples to be photographed, the camera shoots the apples on one side of the black rubber plate, the motor 412 drives the telescopic rod to contract after the apples are shot, the black rubber plate rises, the electric push rod 512 contracts, and the apples are conveyed outwards through the door-shaped frame 41.
Different from the prior art, the technical scheme has the following beneficial effects:
(1) the device provided by the invention comprises an appearance data acquisition device, a data classification device, a data cutting device and a picture generation device which are respectively connected with a controller, and particularly, the appearance data acquisition device has a structure which enables apples to be completely sent to an apple transmission device one by one, so that favorable conditions are provided for the subsequent photographing link.
An output shaft of a motor on the door-shaped frame is connected with a telescopic rod, the telescopic rod is connected with a black rubber plate, and the black rubber plate is driven to act, so that on one hand, the interception of other apples in the slide way is realized, on the other hand, the shooting environment of the apples is unified into a black background plate, and meanwhile, a light supplement lamp is adopted to polish the apples, so that the shooting effect is improved; in addition, before taking a picture, the monitoring video recorder firstly records the position of the apple, and determines whether the position of the apple exposed in the shooting range of the camera is a preset area, if the position of the apple in the groove exposed in the shooting range of the camera is not in the preset area, the controller gives the size of the apple deviating from the preset area, and meanwhile, in order to conveniently adjust the position of the apple, a clamping mechanism is arranged at the lower position of the portal frame, a clamping baffle plate is arranged in the middle of the portal frame, and two ends of the clamping baffle plate are respectively connected with a shifting inclined plate, so that the apple is conveniently limited in position and is conveniently discharged; the accurate adjustment of the position of the apple is achieved through the cooperation between the air faucet and the air pump, so that the apple shooting area is kept uniform and stable, and the acquisition quality of an original data set is improved.
(2) The picture generation method for generating the appearance data of the Fuji apples based on the deep convolution is adopted, when the countermeasure network is trained, one type of pictures is selected each time, the pictures are trained, and the parameters of the generator and the discriminator are continuously updated, so that sample pictures with high quality and complete data of different types of grades are generated more quickly and effectively, and more reference data are provided for apple appearance research.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a schematic diagram illustrating an apparatus for enhancing shape data of red fuji apples according to an embodiment of the present invention;
FIG. 2 is a flowchart of a method for enhancing shape data of Fuji apples according to one embodiment of the present invention;
fig. 3 is a schematic structural diagram of an apparatus for enhancing shape data of red fuji apples according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of a cylindrical feeding hopper of a device for enhancing shape data of Fuji apples according to an embodiment of the present invention;
fig. 5 is a schematic view of a clamping mechanism of a device for enhancing outline data of red fuji apples according to an embodiment of the invention;
FIG. 6 is a comparison of an original sample of the appearance of red Fuji apples and a generated image in accordance with one embodiment of the present invention;
FIG. 7 is a schematic diagram of a generated image of the outline of a red Fuji apple in accordance with one embodiment of the present invention;
FIG. 8 is a diagram illustrating a deep convolution generated countermeasure network according to one embodiment of the present invention;
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
As shown in fig. 1, an embodiment of the present invention provides a device for enhancing appearance data of red fuji apples, including:
the appearance data acquisition device 101 is used for photographing the apples and forming an original data set by the photographed pictures;
the data classification device 102 is configured to classify and calibrate the pictures in the original data set to obtain a calibrated data set;
the data cutting device 103 is used for defining an interested region for the pictures in the calibration data set, cutting and reserving the interested region, performing median filtering on the interested region, and forming a sample data set by the pictures stored in the interested region after noise is eliminated;
the picture generation device 104 is used for inputting the sample data set into a deep convolution generation countermeasure network, performing training processing, generating a new apple picture and forming a result data set;
a controller 105 for controlling the functions of the shape data acquiring device 101, the data sorting device 102, the data cropping device 103 and the picture generating device 104, and storing the data in the execution process of the devices in the memory of the controller.
Further, as shown in fig. 3, the shape data obtaining device 101 includes: the device comprises a frame 1, an apple transmission device, an apple feeding mechanism, an appearance collecting device and an apple position adjusting device;
the apple transmission device is arranged on the upper surface of the frame 1 and extends along the upper surface of the frame 1;
the apple feeding mechanism is arranged at one end of the apple transmission device and stretches across the upper part of the apple transmission device;
the appearance acquisition device is arranged at the other end of the apple transmission device and stretches across the upper part of the apple transmission device;
the apple position adjusting device is arranged on the apple transmission device and is positioned below the appearance acquisition device.
Further, as shown in fig. 3, the apple gear comprises: a conveyor belt 21, a barrier rib 22;
the conveyor belt 21 is positioned on the upper surface of the rack 1, and the conveying direction of the conveyor belt 21 is that the conveyor belt translates from the apple feeding mechanism to the shape acquisition device along the upper surface of the rack 1;
the stop bars 22 are vertically arranged on the upper surface of the frame 1 and are close to two side edges of the conveyor belt 21; the distance between the two barrier strips 22 is greater than the maximum diameter of the apple;
wherein the conveyor belt 21 includes: a recess 211;
the grooves 211 are used for containing apples, are hemispherical, and are uniformly arranged on the conveyor belt 21 along the conveying direction of the conveyor belt 21;
the groove 211 includes: a vent 2111;
the vent 2111 is arranged in the groove 211 and used for blowing air to the apple through the vent to finely adjust the position of the apple.
Further, as shown in fig. 3 and 4, the apple feeding mechanism includes: the cylindrical feed hopper 31 is arranged at one end of the apple transmission device opposite to the appearance acquisition device, and the cylindrical feed hopper 31 stretches over the apple transmission device; the cylindrical feed hopper 31 may be in the shape of a cubic cylinder or a cylinder;
a cylindrical feed hopper as shown in fig. 4, the cylindrical feed hopper 31 comprising: a stepping motor 311, a rotating shaft 312, a dispersing air bag 313, a rotating impeller 314, and a through hole 315;
wherein the content of the first and second substances,
the top of the cylindrical feed hopper 31 is provided with a feed inlet;
the stepping motor 311 is installed at the upper side inside the cylindrical feed hopper 31;
one end of the rotating shaft 312 is connected to the stepping motor 311, and the other end of the rotating shaft 312 extends toward the bottom of the cylindrical hopper 31;
the dispersion air bag 313 is connected to one end of the rotating shaft 312 near the bottom of the cylindrical feed hopper 31; the dispersing air bag 313 is used for preliminarily dispersing the apples poured from the feeding hole into the rotating impeller 314;
the number of the rotating impellers 314 is three, the rotating impellers 314 may be cylindrical boxes, and the three rotating impellers 314 are uniformly distributed and fixed at one end of the rotating shaft 312 close to the bottom of the cylindrical feed hopper 31 and are located below the dispersing air bags 313; an apple outlet is formed in the side wall of the rotary impeller 314;
the through-hole 315 is located the bottom of cylindricality feeder hopper 31 is close to on the lateral wall of appearance collection system one side, just the center of through-hole 315 with on the conveyer belt 21 the line of central connecting of recess 211 aligns, and works as rotatory impeller 314 rotates to just right when the through-hole 315, the center of through-hole 315 with the center of apple export aligns.
Further, as shown in fig. 3, the shape acquisition apparatus includes: a gate frame 41;
the door-shaped frame 41 spans the stop strip 22;
the door-shaped frame 41 is provided with a camera 411, a motor 412, a fill-in light 413 and a monitoring video recorder 414;
the camera 411 is used for taking a picture of the apple;
an output shaft of the motor 412 is connected with a telescopic rod; the telescopic rod is connected with a black rubber plate; the telescopic rod is used for adjusting the position of the black rubber plate under the driving of the motor 412; the black rubber plate is used for providing a black background and intercepting the apples to be photographed subsequently on the conveyor belt 21 when the camera 411 takes photos for the apples;
the fill light 413 is used for providing ambient brightness when the camera 411 shoots;
the monitoring recorder 414 is used for recording the apple to be photographed to obtain the position of the apple to be photographed exposed to the shooting range of the camera 411.
Further, the apple position adjusting device includes: a clamping mechanism and an air tap;
the clamping mechanism is positioned below the appearance acquisition device;
as shown in fig. 5, the clamping mechanism includes: two clamping baffles 5111, four shifting inclined plates 5112 and two electric push rods 512;
the two clamping baffles 5111 are symmetrically arranged on two sides of the central connecting line of the groove 211 on the conveyor belt 21 in parallel and are positioned on the inner sides of the two stop strips 22;
the four shifting inclined plates 5112 are respectively connected to two ends of the two clamping baffles 5111, one end of any shifting inclined plate 5112 is connected to the clamping baffle 5111, and the other end of the shifting inclined plate 5112 extends in a direction obliquely adjacent to the stop strip 22;
one end of each electric push rod 512 is connected with one clamping baffle 5111, and the other end of each electric push rod 512 penetrates through the stop strip 22 adjacent to the clamping baffle 5111 and is installed on the rack 1 through a connecting seat;
the air nozzle is positioned below the conveyor belt 21 and is aligned with a central connecting line of the groove 211 on the conveyor belt 21; the air tap is connected with the air pump; an electromagnetic control valve is arranged on a connecting pipeline of the air tap and the air pump; the solenoid control valves are controlled by the controller 105; the air tap is used for blowing air to the apple through the vent 2111 to finely adjust the position of the apple.
Further, the data classification device 102 specifically includes:
the processing unit of the controller 105 identifies and extracts texture and shape information of the pictures in the original data set, divides the pictures in the original data set into three types of top-class fruits, first-class fruits and second-class fruits, and respectively calibrates the divided three types of apple pictures.
Further, the picture generation apparatus 104 includes: a generator and a discriminator;
the generator is used for performing upsampling by adopting fractional step convolution;
the discriminator is used for performing downsampling by adopting integer step length convolution;
the picture generating device 104 is specifically configured to:
selecting a type of pictures subjected to noise reduction processing and inputting the pictures into a generator, wherein the generator generates a type of new pictures; simultaneously inputting the pictures subjected to noise reduction processing and the generated new pictures into the discriminator; fixing the generator, training the discriminator, stopping training the discriminator when the output value of the corresponding loss function reaches a preset convergence value, further fixing the discriminator, training the generator, stopping training the generator when the output value of the corresponding loss function reaches the preset convergence value, and repeating the steps until the generator and the discriminator model reach a Nash equilibrium state by continuously updating the parameters of the generator and the discriminator.
Example 1:
before explaining the embodiments of the present invention in detail, it is stated that fig. 6, fig. 7, and fig. 8 are drawings for understanding the working process of the deep convolution generation countermeasure network in the picture generation apparatus 104 according to the embodiments of the present invention, which includes the apple-shaped features, but is not limited to the range of the features of the deep convolution generation countermeasure network according to the embodiments of the present invention, and the features of the deep convolution generation countermeasure network according to the embodiments of the present invention may also include, but are not limited to: the color, fruit shape, fruit rust and fruit surface defects of the apples in the apple photos and the characteristics which are self-identified by a countermeasure network algorithm generated by deep convolution and cannot be distinguished by human beings;
beginning with the description of the embodiments of the present invention, as shown in fig. 1, the apparatus for enhancing appearance data of red fuji apples according to the embodiments of the present invention includes: an outline data acquisition means 101, a data sorting means 102, a data cropping means 103, a picture generation means 104, and a controller 105; the appearance data acquisition device 101, the data classification device 102, the data cropping device 103 and the picture generation device 104 are respectively connected with the controller 105;
the shape data acquisition device 101 is used for photographing the apples to acquire an original data set; as shown in fig. 3, the shape data acquisition device comprises a frame 1, an apple feeding mechanism, an apple conveying device, a shape acquisition device and an apple position adjusting device; an apple transmission device is arranged on the upper surface of the frame 1; the apple feeding mechanism is positioned at one end of the apple transmission device and stretches across the apple transmission device; the apple feeding mechanism is used for pouring apples into the apple feeding mechanism and dispersedly and gradually releasing the apples into the grooves 211 of the conveyor belt 21 of the apple transmission device; as shown in fig. 4, the apple feeding mechanism includes a cylindrical hopper 31, a stepping motor 311, a rotating shaft 312, a dispersing air bag 313, a rotating impeller 314, and a through hole 315; the top of the cylindrical feed hopper 31 is provided with a feed inlet; the apples are poured into a cylindrical feed hopper 31 through a feed inlet; the stepping motor 311 is arranged at the upper side inside the cylindrical feed hopper 31; one end of the rotating shaft 312 is connected to the stepping motor 311, and the other end of the rotating shaft 312 extends toward the bottom of the cylindrical hopper 31; the dispersing air bag 313 is connected to one end of the rotating shaft 312 near the bottom of the cylindrical hopper 31; the dispersing air bag 313 is used for preliminarily dispersing the apples poured from the feeding hole into the rotary impeller 314 to avoid the apples from being stacked; three rotating impellers 314 are uniformly distributed and fixed at one end of the rotating shaft 312 close to the bottom of the cylindrical feed hopper 31 and positioned below the dispersing air bag 313; the stepping motor 311 drives the rotary impeller 314 to slowly rotate through the rotary shaft 312 under the control of the controller 105; the rotary impeller 314 can be a cylindrical box body, the apples are dispersed into the three rotary impellers 314 through the dispersing air bags 313, and apple outlets are formed in the side walls of the rotary impellers 314; the through hole 315 is positioned on the side wall of the bottom of the cylindrical feed hopper 31 close to one side of the shape collecting device, the center of the through hole 315 is aligned with the central connecting line of the groove 211 on the conveyor belt 21, and when the rotary impeller 314 rotates to be opposite to the through hole 315, the center of the through hole 315 is aligned with the center of the apple outlet; when any rotating impeller 314 rotates the apple carried by the apple until the apple outlet is aligned with the through hole 315, the apple falls from the cylindrical hopper 31 into the groove 211 of the conveyor belt 21; the apple transmission device comprises a conveyor belt 21 and a barrier strip 22; a row of grooves 211 are uniformly formed in the conveyor belt 21 along the conveying direction of the conveyor belt, the grooves 211 are used for containing apples, so that the apples are conveyed forwards stably, and air vents 2111 are uniformly formed in the grooves 211; two side edges of the frame 1 close to the conveyor belt are respectively provided with a stop strip 22; the distance between the two barrier ribs 22 is slightly larger than the diameter of the apple, specifically, the distance between the two barrier ribs 22 may be the maximum diameter of the apple obtained by a large number of statistics of the diameter of the apple, plus a margin of, for example, 0.5 cm to 10 cm; the apples are placed in the recesses 211 and as the conveyor belt 21 moves towards the profile collection means, the two blocking bars 22 prevent the apples from leaving the correct position when falling from the cylindrical hopper 31 onto the conveyor belt 21, ensuring that the apples fall into the recesses 22. The other end of the apple transmission device, opposite to the apple feeding mechanism, on the apple transmission device is provided with an appearance acquisition device, and the appearance acquisition device comprises a door-shaped frame 41; a motor 412, a light supplement lamp 413, a monitoring video recorder 414 and a camera 411 are arranged on the beam of the portal frame 41; the door-shaped frame 41 spans the stop strip 22; the camera 411 is used to take a picture of the apple; an output shaft of the motor 412 is connected with a telescopic rod; the telescopic rod is connected with a black rubber plate; the telescopic rod is used for adjusting the position of the black rubber plate under the driving of the motor 412; the black rubber plate is used for providing a black background and intercepting the apples to be photographed subsequently on the conveyor belt 21 when the camera 411 takes a picture for the apples; the fill light 413 is used to provide ambient brightness when the camera 411 shoots; the monitoring recorder 414 is used for recording the apple to be photographed to obtain the position of the apple to be photographed exposed to the shooting range of the camera 411. Be provided with apple position adjustment device between appearance collection system's below and conveyer belt 21, include: a clamping mechanism and an air tap; the clamping mechanism is positioned below the appearance acquisition device; the clamping mechanism is arranged to conveniently adjust the position of the apple; the fixture includes: two clamping baffles 5111, four shifting inclined plates 5112 and two electric push rods 512; the two clamping baffles 5111 are symmetrically arranged on two sides of a central connecting line of the groove 211 on the conveyor belt 21 in parallel and are positioned on the inner side of the barrier strip 22; the four shifting inclined plates 5112 are respectively connected to two ends of the two clamping baffles 5111, one end of each shifting inclined plate 5112 is connected to the clamping baffle 5111, and the other end of each shifting inclined plate 5112 extends in a direction obliquely adjacent to the stop strip 22; one end of each electric push rod 512 is connected with a clamping baffle 5111, and the other end of each electric push rod 512 penetrates through the stop strip 22 adjacent to the clamping baffle 5111 and is installed on the rack 1 through a connecting seat; the air nozzle is positioned below the conveyor belt 21 and is aligned with a central connecting line of the groove 211 on the conveyor belt 21; the air nozzle is connected with the air pump; an electromagnetic control valve is arranged on a connecting pipeline of the air tap and the air pump; the solenoid control valves are controlled by the controller 105; the air nozzle is used for blowing air to the apple through the vent 2111 so as to finely adjust the position of the apple. When the apples in the grooves 211 on the conveyor belt 21 pass through the lower end of the portal frame 41, the controller 105 controls the output shaft of the motor 412 to drive the telescopic rod to extend, the telescopic rod drives the black rubber plate to move downwards, and meanwhile, the electric push rod 512 adjusts the telescopic length of the electric push rod 512 according to the diameter of the apples conveyed to the lower end of the portal frame 41, so that the position of the apples is limited; the controller 105 controls the monitoring video recorder 414 to record the position of the apple, the acquired data is uploaded to the controller 105, the controller 105 compares the acquired data with preset data in the controller 105 to determine whether the position of the apple exposed in the shooting range of the camera 411 is a preset area, if the position of the apple exposed in the shooting range of the camera 411 in the groove 211 is not in the preset area, the controller 105 gives the size of the apple deviating from the preset area, meanwhile, the controller 105 controls an air nozzle at the bottom of the corresponding conveyor belt 21 to work, fine adjustment is carried out on the position of the apple until the shooting area of the apple is matched with the preset area, and the air nozzle is closed; meanwhile, the controller 105 adjusts the brightness of the light supplement lamp 413 to supplement light for the apples to be photographed, the camera shoots the apples on one side of the black rubber plate, after shooting is completed, the motor 412 drives the telescopic rod to contract, the black rubber plate rises, the electric push rod 512 contracts, and the apples are conveyed outwards through the door-shaped frame 41;
the raw data set acquired by the profile data acquisition means 101 is stored in the storage unit of the controller 105; the data classification device 102 is used for classifying data in the original data set; the controller 105 controls the data classification device 102 to identify and extract texture and shape information of the pictures of the original data set, takes standard pictures issued by NYT 439-2001-apple appearance grade standard as classification bases, divides the pictures of the original data set into three types of special fruits, first fruits and second fruits, and respectively calibrates the three types of well-divided apple pictures;
the controller 105 controls the data cutting device 103 to divide the region of interest of the calibrated apple picture and cut the region of interest to obtain a cut picture; performing noise reduction processing on the cut picture by using median filtering; specifically, the obtained pictures are processed in a batch, so that each shot picture has the same size and pixels; meanwhile, dividing the pictures of the original data set into three types of special fruits, first fruits and second fruits, and respectively calibrating the divided three types of apple pictures; cutting the region of interest of the calibrated apple picture; carrying out noise reduction processing on the cut picture;
the controller 105 continues to control the picture generation device 104, inputs the picture subjected to the noise reduction processing into a deep convolution generation countermeasure network, performs training processing, and generates a new apple picture; as shown in fig. 8, the deep convolution generation countermeasure network is divided into a generator and a discriminator; the generator performs up-sampling by adopting fractional step convolution, and the discriminator performs down-sampling by adopting integer step convolution; the generator G is generated in a mode of dual with the discriminator D; building the generator G by using a BN algorithm; inputting the picture subjected to noise reduction processing into a generator; the specific network structure of the generator is that 1 x 100 vector is input, the vector is converted into 4 x 1024 through a full connection layer, and an image is generated through 4 up-sampling deconvolution networks; if the left column of fig. 6 is a sample in the original data set, the right column of fig. 6 is a sample of the generated image corresponding to the left side; the specific network structure of the discriminator is that a graph with 64 × 64 size is input, and the graph undergoes 4 times of convolution, and the resolution is reduced to 4 × 4; the specific training steps of the generator and the discriminator are to select a class of pictures after noise reduction processing to be input into the generator, and the generator generates a class of new pictures; simultaneously inputting the pictures subjected to noise reduction processing and the generated new pictures into the discriminator; firstly fixing a generator, training a discriminator, stopping training the discriminator when the output value of the corresponding loss function reaches a preset convergence value, further fixing the discriminator, training the generator, stopping training the generator when the output value of the corresponding loss function reaches the preset convergence value, and continuously updating the parameters of the generator and the discriminator in such a circulating way until the models of the generator and the discriminator reach a Nash equilibrium state.
Under the control of the controller 105, the raw data is acquired by the shape data acquiring device 101, the raw data is classified and calibrated by the data classifying device 102, the calibrated data is cut and noise is reduced by the data cutting device 103, and the picture generating device 104 trains and generates a new apple picture, so as to obtain an expanded apple picture result data set shown in fig. 7.
On the other hand, as shown in fig. 2, an embodiment of the present invention provides a method for enhancing shape data of a red fuji apple by using any one of the red fuji apple shape data enhancing apparatuses, including the steps of:
and 204, inputting the sample data set into a deep convolution generation countermeasure network, performing training processing, and generating a new apple picture to form a result data set.
Pouring the apples into an equipment device obtained by any one of the red Fuji apple shape data enhancement devices in the embodiment of the invention, and taking pictures of the apples by using the equipment device through step 201 to form an original data set by the taken pictures; continuing to step 202, classifying and calibrating the pictures in the original data set to obtain a calibration data set; continuing to step 203, defining an interested region for the picture in the calibration data set, cutting and reserving the interested region, performing median filtering on the interested region, and after eliminating noise, forming a sample data set by the picture stored in the interested region; continuing to step 204, inputting the sample data set into a deep convolution generation countermeasure network, performing training processing, and generating a new apple picture to form a result data set; this resulting data set serves as an augmentation to the original data set for further data analysis processing.
Further, the apple is photographed, and the photographed pictures form an original data set, specifically:
the apples are poured into the feeding hole, the controller 105 controls the stepping motor 311 to drive the rotary impeller 314 to rotate slowly, the apples are dispersed by the dispersing air bags 313 and enter the rotary impeller 314, the apples enter the grooves 211 on the conveyor belt 21 one by one through the apple outlets formed in the side wall of the rotary impeller 314 and the through holes 315 in the cylindrical feed hopper 31, when the apples in the grooves 211 on the conveyor belt 21 pass through the lower end of the portal frame 41, the controller 105 controls the output shaft of the motor 412 to drive the telescopic rod to extend, the telescopic rod drives the black rubber plate to move downwards, and meanwhile, the electric push rod 512 adjusts the telescopic length of the electric push rod 512 according to the diameter of the apples conveyed to the lower end of the portal frame 41, so that the position of the apples is limited; the controller 105 controls the monitoring video recorder 414 to record the position of the apple, the obtained data is uploaded to the controller 105, the controller 105 compares the obtained data with preset data in the controller 105 to determine whether the position of the apple exposed in the shooting range of the camera 411 is a preset area, if the position of the apple in the groove 211 exposed in the shooting range of the camera 411 is not in the preset area, the controller 105 gives the size of the apple deviating from the preset area, meanwhile, the controller 105 controls an air nozzle at the bottom of the corresponding conveyor belt 21 to work, fine adjustment is carried out on the position of the apple until the shooting area of the apple is matched with the preset area, and the air nozzle is closed; meanwhile, the controller 105 adjusts the brightness of the light supplement lamp 413 to supplement light for the apples to be photographed, the camera 411 shoots the apples on one side of the black rubber plate, the motor 412 drives the telescopic rod to contract after the apples are shot, the black rubber plate rises, the electric push rod 512 contracts, and the apples are conveyed outwards through the door-shaped frame 41.
The method for enhancing the shape data of the red fuji apple in the embodiment of the present invention is based on any method of the device for enhancing the shape data of the red fuji apple in the embodiment of the present invention, so that a person skilled in the art can understand the method for enhancing the shape data of the red fuji apple in the embodiment of the present invention through an embodiment of the device for enhancing the shape data of the red fuji apple, and details are not described herein.
It should be understood that the specific order or hierarchy of steps in the processes disclosed is an example of exemplary approaches.
Based upon design preferences, it is understood that the specific order or hierarchy of steps in the processes may be rearranged without departing from the scope of the present disclosure. The accompanying method claims present elements of the various steps in a sample order, and are not intended to be limited to the specific order or hierarchy presented.
In the foregoing detailed description, various features are grouped together in a single embodiment for the purpose of streamlining the disclosure. This method of disclosure is not to be interpreted as reflecting an intention that the claimed embodiments of the subject matter require more features than are expressly recited in each claim. Rather, as the following claims reflect, invention lies in less than all features of a single disclosed embodiment. Thus, the following claims are hereby expressly incorporated into the detailed description, with each claim standing on its own as a separate preferred embodiment of the invention.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. To those skilled in the art; various modifications to these embodiments will be readily apparent, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the disclosure. Thus, the present disclosure is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
What has been described above includes examples of one or more embodiments. It is, of course, not possible to describe every conceivable combination of components or methodologies for purposes of describing the aforementioned embodiments, but one of ordinary skill in the art may recognize that many further combinations and permutations of various embodiments are possible. Accordingly, the embodiments described herein are intended to embrace all such alterations, modifications and variations that fall within the scope of the appended claims. Furthermore, to the extent that the term "includes" is used in either the detailed description or the claims, such term is intended to be inclusive in a manner similar to the term "comprising" as "comprising" is interpreted when employed as a transitional word in a claim. Furthermore, any use of the term "or" in the specification of the claims is intended to mean a "non-exclusive or".
Those of skill in the art will further appreciate that the various illustrative logical blocks, units, and steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both. To clearly illustrate the interchangeability of hardware and software, various illustrative components, elements, and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design requirements of the overall system. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present embodiments.
The various illustrative logical blocks, or elements, described in connection with the embodiments disclosed herein may be implemented or performed with a general purpose processor, a digital signal processor, an Application Specific Integrated Circuit (ASIC), a field programmable gate array or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. A general-purpose processor may be a microprocessor, but in the alternative, the processor may be any conventional processor, controller, microcontroller, or state machine. A processor may also be implemented as a combination of computing devices, e.g., a digital signal processor and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a digital signal processor core, or any other similar configuration.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module may be stored in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. For example, a storage medium may be coupled to the processor such the processor can read information from, and write information to, the storage medium. In the alternative, the storage medium may be integral to the processor. The processor and the storage medium may reside in an ASIC, which may be located in a user terminal. In the alternative, the processor and the storage medium may reside in different components in a user terminal.
In one or more exemplary designs, the functions described above in connection with the embodiments of the invention may be implemented in hardware, software, firmware, or any combination of the three. If implemented in software, the functions may be stored on or transmitted over as one or more instructions or code on a computer-readable medium. Computer-readable media includes both computer storage media and communication media that facilitate transfer of a computer program from one place to another. Storage media may be any available media that can be accessed by a general purpose or special purpose computer. For example, such computer-readable media can include, but is not limited to, RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to carry or store program code in the form of instructions or data structures and which can be read by a general-purpose or special-purpose computer, or a general-purpose or special-purpose processor. Additionally, any connection is properly termed a computer-readable medium, and, thus, is included if the software is transmitted from a website, server, or other remote source via a coaxial cable, fiber optic cable, twisted pair, Digital Subscriber Line (DSL), or wirelessly, e.g., infrared, radio, and microwave. Such discs (disk) and disks (disc) include compact disks, laser disks, optical disks, DVDs, floppy disks and blu-ray disks where disks usually reproduce data magnetically, while disks usually reproduce data optically with lasers. Combinations of the above may also be included in the computer-readable medium.
The above-mentioned embodiments are intended to illustrate the objects, technical solutions and advantages of the present invention in further detail, and it should be understood that the above-mentioned embodiments are merely exemplary embodiments of the present invention, and are not intended to limit the scope of the present invention, and any modifications, equivalent substitutions, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (10)
1. A red Fuji apple appearance data reinforcing means, its characterized in that includes:
the appearance data acquisition device is used for photographing the apples and forming an original data set by the photographed pictures;
the data classification device is used for classifying and calibrating the pictures in the original data set to obtain a calibration data set;
the data cutting device is used for dividing an interested area for the picture in the calibration data set, cutting and reserving the interested area, performing median filtering on the interested area, and after noise is eliminated, forming a sample data set by the picture stored in the interested area;
the picture generating device is used for inputting the sample data set into a deep convolution generation countermeasure network, training and generating a new apple picture to form a result data set;
and the controller is used for controlling the functions of the appearance data acquisition device, the data classification device, the data cutting device and the picture generation device and storing data in the execution process of the devices in a memory of the controller.
2. The apparatus of claim 1, wherein the shape data acquiring means comprises: the apple feeding device comprises a rack (1), an apple transmission device, an apple feeding mechanism, an appearance collecting device and an apple position adjusting device;
the apple transmission device is arranged on the upper surface of the rack (1) and extends along the upper surface of the rack (1);
the apple feeding mechanism is arranged at one end of the apple transmission device and stretches across the upper part of the apple transmission device;
the appearance acquisition device is arranged at the other end of the apple transmission device and stretches across the upper part of the apple transmission device;
the apple position adjusting device is arranged on the apple transmission device and is positioned below the appearance acquisition device.
3. The apparatus as claimed in claim 2, wherein the apple drive means comprises: a conveyor belt (21) and a barrier strip (22);
the conveyor belt (21) is positioned on the upper surface of the rack (1), and the conveying direction of the conveyor belt (21) is that the conveyor belt translates from the apple feeding mechanism to the shape acquisition device along the upper surface of the rack (1);
the stop strips (22) are vertically arranged on the upper surface of the frame (1) and abut against two side edges of the conveyor belt (21);
wherein the conveyor belt (21) comprises: a recess (211);
the grooves (211) are used for containing apples, are hemispherical and are uniformly arranged on the conveyor belt (21) along the conveying direction of the conveyor belt (21);
the groove (211) comprises: a vent (2111);
the vent hole (2111) is disposed in the recess (211).
4. The apparatus of claim 3, wherein the apple feed mechanism comprises: a cylindrical feed hopper (31);
the cylindrical feed hopper (31) comprises: a stepping motor (311), a rotating shaft (312), a dispersing air bag (313), a rotating impeller (314) and a through hole (315);
wherein the content of the first and second substances,
the top of the cylindrical feed hopper (31) is provided with a feed inlet;
the stepping motor (311) is arranged on the upper side inside the cylindrical feed hopper (31);
one end of the rotating shaft (312) is connected with the stepping motor (311), and the other end of the rotating shaft (312) extends towards the bottom of the cylindrical feed hopper (31);
the dispersion air bag (313) is connected to one end of the rotating shaft (312) close to the bottom of the cylindrical feed hopper (31); the dispersing air bag (313) is used for preliminarily dispersing the apples poured from the feeding hole into the rotating impeller (314);
the number of the rotating impellers (314) is three, and the three rotating impellers (314) are uniformly distributed and fixed at one end, close to the bottom of the cylindrical feed hopper (31), of the rotating shaft (312) and are positioned below the dispersing air bag (313); an apple outlet is formed in the side wall of the rotary impeller (314);
through-hole (315) are located the bottom of cylindricality feeder hopper (31) is close to on the lateral wall of appearance collection system one side, just the center of through-hole (315) with on conveyer belt (21) the line of central connecting of recess (211) aligns, and works as rotatory impeller (314) are rotatory to just right when through-hole (315), the center of through-hole (315) with the center of apple export aligns.
5. The red fuji apple shape data enhancement device of claim 3, wherein the shape acquisition means comprises: a gantry (41);
the door-shaped frame (41) spans the barrier strip (22);
a camera (411), a motor (412), a light supplement lamp (413) and a monitoring video recorder (414) are arranged on the door-shaped frame (41);
the camera (411) is used for taking a picture of the apple;
an output shaft of the motor (412) is connected with a telescopic rod; the telescopic rod is connected with a black rubber plate; the telescopic rod is used for adjusting the position of the black rubber plate under the driving of the motor (412); the black rubber plate is used for providing a black background and intercepting apples to be photographed subsequently on the conveyor belt 21 when the camera (411) takes a picture for the apples; (ii) a
The fill light (413) is used for providing ambient brightness when the camera (411) shoots;
the monitoring video recorder (414) is used for carrying out video recording operation on the apple to be photographed so as to obtain the position of the apple to be photographed, which is exposed to the shooting range of the camera (411).
6. The apparatus as claimed in claim 3, wherein the apple position adjusting means comprises: a clamping mechanism and an air tap;
the clamping mechanism is positioned below the appearance acquisition device;
the fixture includes: the device comprises two clamping baffles (5111), four shifting inclined plates (5112) and two electric push rods (512);
the two clamping baffles (5111) are symmetrically arranged on two sides of a central connecting line of the groove (211) on the conveyor belt (21) in parallel and are positioned on the inner sides of the two barrier strips (22);
the four shifting inclined plates (5112) are respectively connected to two ends of the two clamping baffles (5111), one end of any shifting inclined plate (5112) is connected to the clamping baffles (5111), and the other end of the shifting inclined plate (5112) obliquely extends in the direction of the adjacent stop strip (22);
one end of each electric push rod (512) is connected with one clamping baffle (5111), and the other end of each electric push rod (512) penetrates through the stop strip (22) adjacent to the clamping baffle (5111) and is installed on the rack (1) through a connecting seat;
the air nozzle is positioned below the conveyor belt (21) and is aligned with a central connecting line of the groove (211) on the conveyor belt (21); the air tap is connected with the air pump; an electromagnetic control valve is arranged on a connecting pipeline of the air tap and the air pump; the solenoid control valve is controlled by the controller; the air tap is used for blowing air to the apple through the vent hole (2111) to finely adjust the position of the apple.
7. The red Fuji apple shape data enhancement device of claim 1, 2, 3, 4, 5 or 6, wherein the data classification means is specifically:
the processing unit of the controller identifies and extracts the texture and shape information of the pictures in the original data set, divides the pictures in the original data set into three types of special fruits, first fruits and second fruits, and respectively calibrates the divided three types of apple pictures.
8. The apparatus for enhancing the appearance data of Fuji apples according to claim 1, 2, 3, 4, 5 or 6, wherein the picture generating means comprises: a generator and a discriminator;
the generator is used for performing upsampling by adopting fractional step convolution;
the discriminator is used for performing downsampling by adopting integer step length convolution;
the picture generation device is specifically configured to:
selecting a type of pictures subjected to noise reduction processing and inputting the pictures into a generator, wherein the generator generates a type of new pictures; simultaneously inputting the pictures subjected to noise reduction processing and the generated new pictures into the discriminator; fixing the generator, training the discriminator, stopping training the discriminator when the output value of the corresponding loss function reaches a preset convergence value, further fixing the discriminator, training the generator, stopping training the generator when the output value of the corresponding loss function reaches the preset convergence value, and repeating the steps until the generator and the discriminator model reach a Nash equilibrium state by continuously updating the parameters of the generator and the discriminator.
9. A method for enhancing the outline data of red fuji apples, which comprises the steps of using the apparatus for enhancing the outline data of red fuji apples according to any one of claims 1 to 8, comprising:
the apple is photographed, and the photographed pictures form an original data set;
classifying and calibrating the pictures in the original data set to obtain a calibration data set;
defining an interested region for the pictures in the calibration data set, cutting and reserving the interested region, performing median filtering on the interested region, and after eliminating noise, forming a sample data set by the pictures stored in the interested region;
and inputting the sample data set into a deep convolution generation countermeasure network, performing training processing, and generating a new apple picture to form a result data set.
10. The method as claimed in claim 9, wherein the method for enhancing the appearance data of red fuji apples comprises the steps of taking a picture of an apple, and forming an original data set by the taken picture, specifically:
the apples are poured into the feeding hole, the controller (105) controls the stepping motor (311) to drive the rotary impeller (314) to rotate slowly, the apples enter the rotary impeller (314) through the dispersing air bags (313) in a dispersing mode, the apples enter the conveyor belt (21) through the apple outlet formed in the side wall of the rotary impeller (314) and the through hole (315) in the cylindrical feeding hopper (31) one by one, when the apples in the groove (211) in the conveyor belt (21) pass through the lower end of the door-shaped frame (41), the controller (105) controls the output shaft of the motor (412) to drive the telescopic rod to extend, the telescopic rod drives the black rubber plate to move downwards, meanwhile, the electric push rod (512) adjusts the telescopic length of the electric push rod (512) according to the diameter of the apples conveyed to the lower end of the door-shaped frame (41), and therefore the position of the apples is; the controller (105) controls the monitoring video recorder (414) to record the position of the apple, the acquired data are uploaded to the controller (105), the controller (105) compares the acquired data with preset data in the controller (105) to determine whether the position of the apple exposed in the shooting range of the camera (411) is a preset area, if the position of the apple exposed in the shooting range of the camera (411) in the groove (211) is not in the preset area, the controller (105) gives the size of the apple deviating from the preset area, meanwhile, the controller (105) controls an air nozzle at the bottom of the corresponding conveyor belt (21) to work, the position of the apple is finely adjusted until the shooting area of the apple is matched with the preset area, and the air nozzle is closed; meanwhile, the controller (105) adjusts the brightness of the light supplement lamp (413), the apples to be photographed are supplemented with light, the cameras shoot the apples on one side of the black rubber plate, the motor (412) drives the telescopic rod to contract after the apples are shot, the black rubber plate rises, the electric push rod (512) contracts, and the apples are conveyed outwards through the door-shaped frame (41).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011236498.3A CN112560575B (en) | 2020-11-09 | 2020-11-09 | Red Fuji apple shape data enhancement device and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011236498.3A CN112560575B (en) | 2020-11-09 | 2020-11-09 | Red Fuji apple shape data enhancement device and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112560575A true CN112560575A (en) | 2021-03-26 |
| CN112560575B CN112560575B (en) | 2023-07-18 |
Family
ID=75041900
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011236498.3A Active CN112560575B (en) | 2020-11-09 | 2020-11-09 | Red Fuji apple shape data enhancement device and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112560575B (en) |
Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN205926355U (en) * | 2016-06-08 | 2017-02-08 | 江南大学 | Automatic grading system of apple based on apple is directional |
| CN207054735U (en) * | 2017-06-15 | 2018-03-02 | 广东灿城农产品集团有限公司 | A kind of corer for apples |
| CN107862344A (en) * | 2017-12-01 | 2018-03-30 | 中南大学 | A kind of image classification method |
| CN109220194A (en) * | 2018-10-29 | 2019-01-18 | 吕梁学院 | A kind of picking of fruit auxiliary and categorised collection all-in-one machine |
| CN109460708A (en) * | 2018-10-09 | 2019-03-12 | 东南大学 | A kind of Forest fire image sample generating method based on generation confrontation network |
| CN109493308A (en) * | 2018-11-14 | 2019-03-19 | 吉林大学 | The medical image synthesis and classification method for generating confrontation network are differentiated based on condition more |
| CN109794427A (en) * | 2019-02-03 | 2019-05-24 | 东莞市粤蒙电子科技有限公司 | A kind of screening installation |
| CN109894383A (en) * | 2019-02-21 | 2019-06-18 | 南方科技大学 | A kind of article sorting method, device, storage medium and electronic equipment |
| CN110097107A (en) * | 2019-04-23 | 2019-08-06 | 安徽大学 | Alternaria mali roberts disease recognition and classification method based on convolutional neural networks |
| CN110148122A (en) * | 2019-05-17 | 2019-08-20 | 南京东奇智能制造研究院有限公司 | Apple presentation quality stage division based on deep learning |
| CN110188824A (en) * | 2019-05-31 | 2019-08-30 | 重庆大学 | A kind of small sample plant disease recognition methods and system |
| US20190286950A1 (en) * | 2018-03-16 | 2019-09-19 | Ebay Inc. | Generating a digital image using a generative adversarial network |
| CN110276386A (en) * | 2019-06-11 | 2019-09-24 | 济南大学 | A kind of apple grading method and system based on machine vision |
| CN110569864A (en) * | 2018-09-04 | 2019-12-13 | 阿里巴巴集团控股有限公司 | vehicle loss image generation method and device based on GAN network |
| CN111076670A (en) * | 2019-12-03 | 2020-04-28 | 北京京仪仪器仪表研究总院有限公司 | Online nondestructive testing method for internal and external quality of apples |
| CN111739017A (en) * | 2020-07-22 | 2020-10-02 | 湖南国科智瞳科技有限公司 | Cell identification method and system of microscopic image under sample unbalance condition |
-
2020
- 2020-11-09 CN CN202011236498.3A patent/CN112560575B/en active Active
Patent Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN205926355U (en) * | 2016-06-08 | 2017-02-08 | 江南大学 | Automatic grading system of apple based on apple is directional |
| CN207054735U (en) * | 2017-06-15 | 2018-03-02 | 广东灿城农产品集团有限公司 | A kind of corer for apples |
| CN107862344A (en) * | 2017-12-01 | 2018-03-30 | 中南大学 | A kind of image classification method |
| US20190286950A1 (en) * | 2018-03-16 | 2019-09-19 | Ebay Inc. | Generating a digital image using a generative adversarial network |
| CN110569864A (en) * | 2018-09-04 | 2019-12-13 | 阿里巴巴集团控股有限公司 | vehicle loss image generation method and device based on GAN network |
| CN109460708A (en) * | 2018-10-09 | 2019-03-12 | 东南大学 | A kind of Forest fire image sample generating method based on generation confrontation network |
| CN109220194A (en) * | 2018-10-29 | 2019-01-18 | 吕梁学院 | A kind of picking of fruit auxiliary and categorised collection all-in-one machine |
| CN109493308A (en) * | 2018-11-14 | 2019-03-19 | 吉林大学 | The medical image synthesis and classification method for generating confrontation network are differentiated based on condition more |
| CN109794427A (en) * | 2019-02-03 | 2019-05-24 | 东莞市粤蒙电子科技有限公司 | A kind of screening installation |
| CN109894383A (en) * | 2019-02-21 | 2019-06-18 | 南方科技大学 | A kind of article sorting method, device, storage medium and electronic equipment |
| CN110097107A (en) * | 2019-04-23 | 2019-08-06 | 安徽大学 | Alternaria mali roberts disease recognition and classification method based on convolutional neural networks |
| CN110148122A (en) * | 2019-05-17 | 2019-08-20 | 南京东奇智能制造研究院有限公司 | Apple presentation quality stage division based on deep learning |
| CN110188824A (en) * | 2019-05-31 | 2019-08-30 | 重庆大学 | A kind of small sample plant disease recognition methods and system |
| CN110276386A (en) * | 2019-06-11 | 2019-09-24 | 济南大学 | A kind of apple grading method and system based on machine vision |
| CN111076670A (en) * | 2019-12-03 | 2020-04-28 | 北京京仪仪器仪表研究总院有限公司 | Online nondestructive testing method for internal and external quality of apples |
| CN111739017A (en) * | 2020-07-22 | 2020-10-02 | 湖南国科智瞳科技有限公司 | Cell identification method and system of microscopic image under sample unbalance condition |
Non-Patent Citations (3)
| Title |
|---|
| JORDI GENÉ-MOLA 等: "Multi-modal deep learning for Fuji apple detection using RGB-D cameras and their radiometric capabilities", 《COMPUTERS AND ELECTRONICS IN AGRICULTURE 162》, pages 689 - 698 * |
| WEI FANG 等: "A Method for Improving CNN-Based Image Recognition Using DCGAN", 《CMC》, vol. 57, no. 1, pages 167 - 178 * |
| 宋燕星 等: "扫描激光线源技术检测金属表面缺陷位置实验研究", 《计算机测量与控制》, vol. 24, no. 11, pages 37 - 40 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112560575B (en) | 2023-07-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8340452B2 (en) | Automatic generation of a photo guide | |
| DE602006000400T2 (en) | PICTORIALIZATION PROCESS AND SYSTEM | |
| US5220614A (en) | Automated coin grading system | |
| CN102460115B (en) | Flow type particle image analysis method and device | |
| US20020085219A1 (en) | Method of and system for generating and viewing multi-dimensional images | |
| CN111179223A (en) | Deep learning-based industrial automatic defect detection method | |
| CN110479635B (en) | Method and device for automatically sorting tobacco leaves based on neural network | |
| DE112008001990T5 (en) | Exposure control for an imaging system | |
| CN110490252B (en) | Indoor people number detection method and system based on deep learning | |
| CN113624645B (en) | Device and method for detecting particle flow distribution | |
| CN110233971A (en) | A kind of image pickup method and terminal, computer readable storage medium | |
| CN112974303B (en) | Hyperspectrum-based fruit quality detection method, device and medium | |
| CN112560575A (en) | Red Fuji apple shape data enhancement device and method | |
| CA3193037A1 (en) | Camera image or video processing pipelines with neural embedding | |
| CN115862004A (en) | Corn ear surface defect detection method and device | |
| CN111460946A (en) | Image-based chip information rapid acquisition and identification method | |
| CN102369541A (en) | Method for performing automatic classification of image information | |
| CN116912244A (en) | Intelligent grain quality index detection method based on machine vision | |
| CN113296259A (en) | Super-resolution imaging method and device based on aperture modulation subsystem and deep learning | |
| CN112200241A (en) | Automatic sorting method for fish varieties based on ResNet transfer learning | |
| CN108634241A (en) | Select halogen method and its brine-adding system | |
| CN114423531A (en) | Automated particle inspection | |
| Allen et al. | Machine vision for automated optical recognition and classification of pollen grains or other singulated microscopic objects | |
| Cömert et al. | Weight and diameter estimation using image processing and machine learning techniques on apple images | |
| CN114774253A (en) | Portable spore capturing and detecting device and detecting method thereof |
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 |