CN115112508A - Device and method for identifying soft and hard bags of consigned luggage in civil aviation airport - Google Patents
Device and method for identifying soft and hard bags of consigned luggage in civil aviation airport Download PDFInfo
- Publication number
- CN115112508A CN115112508A CN202211036891.7A CN202211036891A CN115112508A CN 115112508 A CN115112508 A CN 115112508A CN 202211036891 A CN202211036891 A CN 202211036891A CN 115112508 A CN115112508 A CN 115112508A
- Authority
- CN
- China
- Prior art keywords
- luggage
- point cloud
- baggage
- sample
- hard
- 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 28
- 238000001125 extrusion Methods 0.000 claims abstract description 35
- 230000007246 mechanism Effects 0.000 claims description 19
- 238000010606 normalization Methods 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 8
- 238000012216 screening Methods 0.000 claims description 7
- 238000003825 pressing Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000003064 k means clustering Methods 0.000 claims description 3
- 230000000007 visual effect Effects 0.000 claims description 2
- 230000008569 process Effects 0.000 abstract description 10
- 230000009466 transformation Effects 0.000 description 7
- 239000011159 matrix material Substances 0.000 description 6
- 239000000463 material Substances 0.000 description 5
- 238000010586 diagram Methods 0.000 description 3
- 238000009432 framing Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 230000003111 delayed effect Effects 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000004744 fabric Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 238000003466 welding Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N3/00—Investigating strength properties of solid materials by application of mechanical stress
- G01N3/40—Investigating hardness or rebound hardness
- G01N3/42—Investigating hardness or rebound hardness by performing impressions under a steady load by indentors, e.g. sphere, pyramid
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N3/00—Investigating strength properties of solid materials by application of mechanical stress
- G01N3/02—Details
- G01N3/06—Special adaptations of indicating or recording means
- G01N3/068—Special adaptations of indicating or recording means with optical indicating or recording means
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2203/00—Investigating strength properties of solid materials by application of mechanical stress
- G01N2203/0058—Kind of property studied
- G01N2203/0076—Hardness, compressibility or resistance to crushing
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2203/00—Investigating strength properties of solid materials by application of mechanical stress
- G01N2203/02—Details not specific for a particular testing method
- G01N2203/06—Indicating or recording means; Sensing means
- G01N2203/0641—Indicating or recording means; Sensing means using optical, X-ray, ultraviolet, infrared or similar detectors
- G01N2203/0647—Image analysis
Landscapes
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Biochemistry (AREA)
- Analytical Chemistry (AREA)
- Chemical & Material Sciences (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The application relates to the technical field of baggage sorting and discloses a device and a method for identifying soft and hard baggage packages of consigned baggage in a civil aviation airport, wherein the device comprises a rack, a press-touch module, a distance measuring module and a vision module; the rack is erected above the baggage conveying equipment; the press-contact module is arranged on the rack and used for contacting and extruding the luggage, so that the luggage generates concave deformation, and the concave extrusion force at the concave position of the luggage is collected; the distance measuring module is arranged on the rack and used for sensing the passing of the luggage and collecting the height distance between the upper surface of the luggage and the lower end of the press contact module; the vision module is arranged on the rack and used for collecting the concave RGB images and the concave point cloud images at the luggage extrusion concave positions. The application solves the problem that automatic identification and classification of the soft and hard luggage can not be automatically completed in the existing luggage consignment process.
Description
Technical Field
The application relates to the technical field of baggage sorting, in particular to a device and a method for identifying soft and hard bags of consigned baggage in a civil aviation airport.
Background
At present, if luggage for airport passengers to consign is soft luggage such as backpacks, cloth bags, woven bags and the like, framing treatment is required to avoid damage in the conveying and sorting processes; when luggage is piled up on the trailer after being sorted, soft luggage and hard luggage need to be piled up in layers, the soft luggage needs to be placed on the upper layer of the hard luggage, the soft luggage is prevented from being crushed, and the luggage is also arranged when being piled up on an airplane. Soft and hard baggage identification is beneficial to optimizing baggage sorting strategies, reducing baggage breakage rate and improving baggage consignment service quality.
At present, the framing of soft luggage, the stacking of luggage on a vehicle and the stacking of luggage on an airplane are basically completed manually, so that the identification of soft and hard luggage is mainly realized by depending on the experience of an operator, and no reliable and mature automatic classification equipment exists. However, in order to realize the automatic processing of the processes of baggage framing, baggage car stacking, baggage airplane stacking, and the like, identification of soft and hard baggage is one of the necessary conditions.
Disclosure of Invention
Based on the technical problems, the application provides a device and a method for identifying soft and hard bags of consignment luggage in a civil aviation airport, and solves the problem that automatic identification and classification of the soft and hard bags of luggage cannot be automatically completed in the existing luggage consignment process.
In order to solve the technical problems, the technical scheme adopted by the application is as follows:
a device for identifying soft and hard bags of consigned luggage in a civil aviation airport comprises a rack, a press touch module, a distance measuring module and a vision module; the rack is erected above the baggage conveying equipment; the press-contact module is arranged on the rack and used for contacting and extruding the luggage, so that the luggage generates concave deformation, and the concave extrusion force at the concave position of the luggage is collected; the distance measuring module is arranged on the rack and used for sensing the passing of the luggage and collecting the height distance between the upper surface of the luggage and the lower end of the press contact module; the vision module is arranged on the rack and used for acquiring concave RGB images and concave point cloud images of the luggage extrusion concave position.
Further, the press-contact module comprises a linear driving mechanism and a contact rod; the linear driving mechanism is vertically arranged on the rack; the contact rod is in driving connection with the movable end of the linear driving mechanism through a pressure sensor.
Further, the linear driving mechanism is a servo electric cylinder.
Further, the pressure touch module still includes the pressure touch control module, the pressure touch module is used for based on the height distance with sunken extrusion force control the falling speed of contact lever specifically includes:
if the height distance is greater than a preset safety distance, the descending speed is a first speed;
if the height distance is smaller than a preset safety distance, the descending speed is a second speed;
if the depression extrusion force is equal to the preset safety pressure, the descending speed is zero;
wherein the first speed is greater than the second speed.
Further, a baggage conveyor control module is included, the baggage conveyor control module comprising:
the size recognition unit is used for calculating the length of the luggage along the conveying direction through the luggage point cloud image acquired by the vision module;
a time control unit for controlling the time of the clock based on the function
Calculating the time interval
Wherein d represents the length of the luggage,
representing a conveying speed of the baggage conveying device;
and the starting and stopping control unit is used for suspending the luggage conveying equipment after the interval time when the ranging module senses that the luggage passes.
A method for identifying soft and hard bags of consigned luggage in a civil aviation airport, which is based on the identification device for soft and hard bags of consigned luggage in the civil aviation airport, comprises the following steps:
acquiring depression data of a luggage squeezing position, wherein the depression data comprises depression squeezing force, a depression RGB image and a depression point cloud image of the luggage squeezing depression;
comparing and screening the luggage data with sample data in a sample set based on a similarity algorithm, wherein the sample data comprises sample extrusion force, a sample RGB image, a sample point cloud image and a sample hardness level;
if the sample data similar to the luggage data is screened out, taking the sample hardness level of the sample data as the luggage hardness level of the luggage;
and if the sample data similar to the luggage data is not screened out, inputting the luggage data into a classification identification model to obtain the luggage soft and hard grade of the luggage.
Further, the step of inputting the baggage data into a classification recognition model to obtain the baggage soft and hard levels of the baggage further includes:
and storing the baggage data and the corresponding baggage soft and hard grades into the sample set.
Further, comparing and screening the luggage data with sample data in a sample set based on a similarity algorithm comprises:
acquiring a difference value between the depression extrusion force and the sample extrusion force, and acquiring a first similarity based on the difference value;
inputting the concave RGB image and the sample RGB image into a Siamese network model to obtain a second similarity;
carrying out similarity calculation on the sunken point cloud image and the sample point cloud image to obtain a third similarity;
and comparing the first similarity, the second similarity and the third similarity with preset similar conditions to judge whether the luggage data is similar to the sample data.
Further, a specific formula for calculating the similarity between the concave point cloud image and the sample point cloud image is as follows:
wherein r (A, B) represents a third similarity, A represents a sunken point cloud image, B represents a sample point cloud image, and the higher the value of r (A, B), the more similar the sunken point cloud image A and the sample point cloud image B are;
the weight of the point is represented by a weight,
the distance of the point cloud is represented,
representing the result obtained by K-means clustering the concave point cloud image AThe ith sub-point cloud of the N sub-point cloud sets,
the point of maximum depth in the vertical direction, i.e. the central position of the push rod,
representing a hyper-parameter;
wherein,
the concrete formula of (1) is as follows:
wherein h represents an adjustable parameter;
wherein,
the concrete formula of (1) is as follows:
wherein,
representing the distance between the point cloud X and the point cloud Y, X representing the points in the point cloud subset X, and Y representing the points in the point cloud subset Y.
Further, the classification identification model comprises a two-dimensional convolution network, a three-dimensional point cloud convolution network, a normalization layer, a vector splicing layer and a full connection layer; inputting the luggage data into a classification recognition model, and obtaining the luggage hardness and softness grade of the luggage comprises the following steps:
inputting the sunken RGB image into the two-dimensional convolution network to obtain a first output result;
inputting the sunken point cloud image into the three-dimensional point cloud convolution network to obtain a second output result;
inputting the first output result, the second output result and the depression extrusion force into the normalization layer for processing, and then inputting the normalization processing result into the vector splicing layer to obtain a splicing vector;
and inputting the splicing vector into the full-connection layer to obtain the luggage soft and hard grade classification result of the luggage.
Compared with the prior art, the beneficial effects of this application are:
this application has synthesized and has utilized power touch control and image recognition technique, touches module contact and presses luggage through pressing, touches and presses the mechanism and takes force sensor certainly, after detecting pressure, examines the RGB picture and some cloud data that acquire luggage deformation with the vision module, through big data analysis and image recognition technique, synthesizes the soft, hard degree of judging luggage, accomplishes soft, hard luggage automatic identification and classification.
The method and the device can automatically and quickly complete the identification and classification of soft and hard luggage, help to optimize subsequent luggage consignment sorting strategies, reduce the breakage rate of the luggage and improve the consignment service quality of the luggage.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the invention without limiting the invention. Wherein:
fig. 1 is a schematic structural diagram of a device for identifying soft and hard bags for consignment luggage in a civil aviation airport.
Fig. 2 is a schematic diagram of the internal structure of the press-contact module.
Fig. 3 is a schematic flow chart of a method for identifying soft and hard baggage of consigned baggage in a civil aviation airport.
Fig. 4 is a schematic flowchart of a comparison and screening process of the baggage data and the sample data in the sample set based on the similarity algorithm.
Fig. 5 is a schematic flow chart of inputting the baggage data into a classification recognition model to obtain the baggage soft and hard levels of the baggage.
FIG. 6 is a block diagram of a classification recognition model.
The system comprises a 1 pressure touch module, a 101 linear driving mechanism, a 102 pressure sensor, a 103 contact rod, a 2 vision module, a 3 rack, 4 luggage, 5 luggage conveying equipment and a 6 distance measuring module.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the drawings of the embodiments of the present application. It should be apparent that the described embodiments are only some of the embodiments of the present application, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the described embodiments of the application without any inventive step, are within the scope of protection of the application.
Unless defined otherwise, technical or scientific terms used herein shall have the ordinary meaning as understood by one of ordinary skill in the art to which this application belongs. As used in this application, the terms "first," "second," and the like do not denote any order, quantity, or importance, but rather are used to distinguish one element from another. The word "comprising" or "comprises", and the like, means that the element or item preceding the word comprises the element or item listed after the word and its equivalent, but does not exclude other elements or items. The terms "connected" or "coupled" and the like are not restricted to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "upper", "lower", "left", "right", and the like are used merely to indicate relative positional relationships, and when the absolute position of the object being described is changed, the relative positional relationships may also be changed accordingly.
Fig. 1 to 2 are schematic structural views of a device for identifying soft and hard bags of checked-in baggage at a civil aviation airport according to some embodiments of the present invention, and the device for identifying soft and hard bags of checked-in baggage at a civil aviation airport according to the present invention will be described with reference to fig. 1 to 2. It should be noted that fig. 1-2 are only examples, and do not limit the specific shape and structure of the baggage identification device for checking baggage at civil aviation airports.
Referring to fig. 1-2, in some embodiments, an identification device for soft and hard baggage of civil aviation airport consignment includes a frame 3, a press touch module 1, a distance measurement module 6 and a vision module 2; the frame 3 is erected above the baggage conveying device 5; the press-contact module 1 is arranged on the rack 3 and used for contacting and pressing the luggage 4, so that the luggage 4 is deformed in a concave manner, and the concave pressing force of the concave position of the luggage 4 is collected; the distance measuring module 6 is arranged on the rack 3 and used for sensing the passing of the luggage 4 and collecting the height distance between the upper surface of the luggage 4 and the lower end of the press-contact module 1; the vision module 2 is arranged on the machine frame 3 and is used for acquiring a concave RGB image and a concave point cloud image of the pressed concave part of the luggage 4.
In the embodiment, when the luggage 4 is conveyed to the front lower part of the luggage 4 soft and hard bag identification device through the luggage 4 conveying mechanism, the distance measuring module 6 on the device detects the luggage 4 and triggers, and collects the height distance between the upper surface of the luggage 4 and the lower end of the press-contact module 1. And at the same time, the baggage 4 conveying mechanism is stopped after a certain time interval is delayed, so that the baggage 4 is stopped right below the device.
The follow-up control is pressed and is touched module 1 and is said carefully and press luggage 4, makes luggage 4 produce sunken deformation to gather the sunken extrusion force of 4 sunken departments of luggage. At this time, a dent appears on the surface of the luggage 4, and the vision module 2 acquires a dent RGB image and a dent point cloud image of the position. And substituting the sunken extrusion force, the sunken RGB image and the sunken point cloud image into a big data analysis and image recognition technology, comprehensively judging the soft and hard degrees of the luggage 4, and finishing the automatic recognition and classification of the soft and hard luggage 4.
In addition, this application has avoided single information to the erroneous judgement of 4 materials of luggage through combining together the sense of force and vision, especially at luggage 4 or parcel material itself, when great with its content material difference, has improved the rate of accuracy of 4 soft and hard discernments of luggage greatly. For example, if the classification of the bag of luggage 4 is determined by merely identifying the material of the bag of luggage 4, it is likely to cause erroneous judgment, such as a canvas bag filled with books or other hard materials, which should also be classified as a hard bag classification.
Specifically, the baggage conveying device 5 is a belt conveying mechanism.
Specifically, the support is a portal frame structure formed by tailor welding and is erected above the belt conveying mechanism, and the lower end of the support is fixedly installed with the frame part of the luggage conveying equipment 5 through bolts or directly installed and fixed with the ground. It is used as the bearing structure of the whole device and is used for installing the press-contact module 1, the distance measuring module 6 and the vision module 2.
Specifically, ranging module 6 is laser range finder, and its hoist and mount is on the lower surface of support crossbeam, and vertical downward carries out the range finding. Since the laser range finder is triggered when the luggage 4 passes through the range finding module 6, it can be determined that the luggage 4 passes through.
Specifically, the vision module 2 is a depth vision camera, and is hung on the lower surface of the support beam. Preferably, to ensure that the depth vision camera can take a picture of the luggage 4 in all directions, two depth vision cameras are symmetrically arranged above the bracket through a hinge mechanism and are inclined inwards to obtain a larger visual range. Since the information of the dent is represented by both the RGB image and the point cloud image, registration of their point clouds is required when there are two depth vision cameras.
The specific process of point cloud registration comprises the steps of establishing a coordinate system at the intersection point of the press-contact module 1 and the support, obtaining the imaging center of the depth vision camera through calibration, and converting point cloud coordinates obtained by the depth vision camera into the coordinate system. In the following processing process, the coordinate positions of the point clouds acquired by the two depth vision cameras are both converted, and the coordinate system is taken as a reference.
Firstly, extracting the characteristic points of the point cloud by using algorithms such as PFH (pulse frequency histogram) and SHOT (SHOT-root), wherein any characteristic point x in the point cloud has the following homogeneous expression:
wherein,
is the i-th dimension value of the feature point x.
Consider that
The projective transformation H in (1), i.e. a 4 × 4 non-singular homogeneous matrix.
Wherein,
the elements of the projective transformation matrix H in the ith row and the jth column are shown,
respectively, homogeneous representations of feature points before and after transformation.
The homogeneous matrix H has 15 degrees of freedom except for a scale factor, 4 pairs or more of characteristic point pairs are substituted in the point clouds acquired by the left and right depth vision cameras to obtain a transformation matrix H, and if the number of the characteristic point pairs is insufficient, the point clouds are acquired repeatedly.
Calculating an optimal transformation matrix H based on RANSAC, specifically comprising:
1. 4 groups of feature corresponding points are randomly selected to calculate a spatial transformation matrix H.
2. Calculating projection errors for all feature points
Wherein
Is a group of characteristic corresponding points, and the characteristic corresponding points,
for the distance function, the difference is usually taken as the two-norm. And if the projection error is smaller than the set threshold value m, adding the point into the inner point set I.
3. If the number of the elements of the current interior point set is larger than the optimal interior point set
Or when the two numbers are the same, and sThe standard deviation of the current interior point set is lower than the optimal interior point set
At the same time, update
Is I.
4. And repeating the process until the number of the points in the inner point set reaches a set threshold value or the iteration number reaches an upper limit, and stopping iterative computation.
For a point in a point cloud
Transformed into another point cloud
Wherein:
and for the image information, converting an original image acquired by the depth vision camera into an image of the horizontal plane of the baggage conveying equipment 5 on a shadow bearing surface by using perspective transformation, registering and splicing the images of the two depth vision cameras, wherein the registering algorithm is similar to point cloud, and extracting the image of the dent part of the baggage 4 according to the color and the frame difference.
Preferably, the press-contact module 1 comprises a linear driving mechanism 101 and a contact rod 103; the linear driving mechanism 101 is vertically arranged on the frame 3; the contact rod 103 is drivingly connected to the movable end of the linear drive mechanism 101 via the pressure sensor 102.
The pressure contact module 1 drives the contact rod 103 to contact or separate from the luggage 4 in the vertical direction through the linear driving mechanism 101, and obtains the pressing force of the contact rod 103 contacting the luggage 4 through the pressure sensor 102.
Specifically, the linear driving mechanism 101 is a servo cylinder. The servo electric cylinder can control the contact rod 103 to extrude the surface of the luggage 4 at a set speed, so that the luggage 4 is deformed, and the accuracy of stroke speed control of the contact rod 103 can be improved.
Preferably, the pressure touch module 1 further includes a pressure touch control module, and the pressure touch control module is configured to control a lowering speed of the contact rod 103 based on the height distance and the depression pressing force, and specifically includes:
if the height distance is greater than the preset safety distance, the descending speed is the first speed;
if the height distance is smaller than the preset safety distance, the descending speed is the second speed;
if the depression extrusion force is equal to the preset safety pressure, the descending speed is zero;
wherein the first speed is greater than the second speed.
Because the first speed is greater than the second speed, when the height distance is greater than the preset safety distance, the contact rod 103 descends at a faster speed, so that the descending time is shortened, and the detection efficiency is improved. When the distance is less than the preset safety distance, i.e. when the contact rod 103 approaches the luggage item 4, it is slowly lowered at a slower speed, in order to avoid that the luggage item 4 is damaged by the impact of the too fast speed on the luggage item 4.
The pressure touch control module controls the set position and speed by a PID algorithm, and the pressure touch control module has position and speed feedback.
And after contact pole 103 and luggage 4 contact, can produce contact pressure, after sunken extrusion force equals to predetermineeing safe pressure, then stop contact pole 103 and push down, avoid the extrusion force too big, destroy luggage 4.
Specifically, the descending speed formula of the contact rod 103 is:
wherein,
indicating the speed of descent of the contact bar,
which is indicative of a first speed of the vehicle,
which is indicative of the second speed of the vehicle,
a preset safety distance is indicated and indicated,
the safety pressure is preset, h represents the height distance, and F represents the extrusion force.
The extrusion force of the pressure contact module 1 adopts a hierarchical extrusion mode, firstly, an initial extrusion force is set to extrude the luggage 4, and if no dent is generated on the surface of the luggage 4 or the dent is very small, the pressure contact module 1 increases the pressure until the set highest preset safety pressure is reached. By setting a variable pressure threshold of the pressure-contact module 1, the safety of the luggage 4 is ensured in case of normal handling recognition of dents.
Preferably, a baggage conveyor device 5 control module is further included, the baggage conveyor device 5 control module including:
the size recognition unit is used for calculating the length of the luggage 4 along the conveying direction through the luggage 4 point cloud image acquired by the vision module 2;
a time control unit for controlling the time of the clock based on the function
Calculating the time interval
Wherein, d represents the length of the luggage,
representing a conveying speed of the baggage conveying device;
and the starting and stopping control unit is used for pausing the luggage conveying equipment 5 after the interval time when the distance measuring module 6 senses that the luggage 4 passes through.
Wherein, the control module of the baggage conveying device 5 obtains the interval time based on the length of the baggage 4 and the conveying speed, and through the interval time, when the baggage 4 passes through the apparatus, the baggage conveying device 5 is suspended after the interval time is delayed, so that the central part of the baggage 4 can be positioned right below the apparatus, thereby facilitating the pressure touch detection of the subsequent pressure touch module 1.
For the length d of the luggage 4, the vision module 2 acquires an image of the current luggage 4, acquires the pixel difference between the upper end point and the lower end point of the luggage 4 according to the color and the frame difference, and obtains the actual length of the unit pixel at the fixed depth by combining the calibration information put into the vision module 2, thereby acquiring the length d of the luggage 4 in the horizontal plane and the conveyor direction.
Referring to fig. 3, a method for identifying hard and soft bags of consignment baggage at a civil aviation airport, based on the identification device for hard and soft bags of consignment baggage at a civil aviation airport, includes:
s301, acquiring depression data of a luggage squeezing position, wherein the depression data comprises depression squeezing force, a depression RGB image and a depression point cloud image of the luggage squeezing depression position;
specifically, the depression extrusion force is acquired through a press-contact module;
specifically, the concave RGB image and the concave point cloud image are acquired through a vision module.
S302, comparing and screening the luggage data with sample data in a sample set based on a similarity algorithm, wherein the sample data comprises sample extrusion force, a sample RGB image, a sample point cloud image and a sample hardness grade;
the sample data contains soft and hard grade information, so that if the extrusion force of the luggage data, the RGB image and the point cloud image are similar to the corresponding information of the sample data, the luggage data is similar to the sample data, and the soft and hard grades of the luggage data and the sample data are similar.
S303, if sample data similar to the luggage data is screened out, taking the sample hardness level of the sample data as the luggage hardness level of the luggage;
the classification recognition model is a neural network model, a large amount of operation resources are consumed for operation, and before the classification recognition model is recognized, the soft and hard recognition efficiency can be improved, the recognition time can be shortened, and the operation resources can be saved by obtaining the soft and hard grades through similarity calculation.
S304, if the sample data similar to the luggage data is not screened out, inputting the luggage data into a classification identification model to obtain the luggage soft and hard grade of the luggage.
Preferably, after the baggage data is input into the classification recognition model and the baggage soft and hard grades of the baggage are obtained, the method further includes: and storing the baggage data and the corresponding baggage soft and hard grades into the sample set.
If the sample data similar to the luggage data is not screened out, it is indicated that no sample data similar to the luggage data exists in the sample set, and the luggage data and the corresponding luggage soft and hard grades can be stored in the sample set, so that the similarity comparison of other luggage data in the follow-up process is facilitated. By the operation, the data volume in the sample set can be increased, so that the sample data contained in the sample set is more and more comprehensive.
Referring to fig. 4, preferably, the comparing and screening the baggage data with the sample data in the sample set based on the similarity algorithm includes:
s401, obtaining a difference value between the depression extrusion force and the sample extrusion force, and obtaining a first similarity based on the difference value;
wherein, for the extrusion force, the numerical difference between the two can be directly compared, and the smaller the difference value, the higher the first similarity of the depression extrusion force and the sample extrusion force.
S402, inputting the concave RGB image and the sample RGB image into a Siamese network model to obtain a second similarity;
wherein, for the siemese network model, which has two shared image input channels, the smaller the value of its output, the higher the second similarity of the recess RGB image and the sample RGB image.
S403, performing similarity calculation on the sunken point cloud image and the sample point cloud image to obtain a third similarity;
s404, comparing the first similarity, the second similarity and the third similarity with preset similar conditions to judge whether the luggage data is similar to the sample data.
Preferably, a specific formula for calculating the similarity between the concave point cloud image and the sample point cloud image is as follows:
wherein r (A, B) represents a third similarity, A represents a sunken point cloud image, B represents a sample point cloud image, and the higher the value of r (A, B), the more similar the sunken point cloud image A and the sample point cloud image B are;
the weight of the point is represented by a weight,
the distance of the point cloud is represented,
representing the ith sub-point cloud in the N sub-point cloud sets obtained by K-means clustering the concave point cloud image A,
the point of maximum depth in the vertical direction, i.e. the central position of the push rod,
representing a hyper-parameter;
in particular, the method comprises the following steps of,
determined as a function of the circumstances, typically being a small positive value, e.g.
Can be arranged as
。
Wherein,
the concrete formula of (2) is as follows:
wherein h represents an adjustable parameter;
wherein,
the concrete formula of (2) is as follows:
wherein,
representing the distance between the point cloud X and the point cloud Y, X representing the points in the point cloud subset X, and Y representing the points in the point cloud subset Y.
Referring to fig. 6, preferably, the classification recognition model includes a two-dimensional convolution network, a three-dimensional point cloud convolution network, a normalization layer, a vector splicing layer and a full connection layer.
Based on the classification recognition model of fig. 6, referring to fig. 5, inputting the baggage data into the classification recognition model, and obtaining the baggage soft and hard levels of the baggage comprises:
s501, inputting the sunken RGB image into the two-dimensional convolution network to obtain a first output result;
specifically, the two-dimensional convolutional network is a Conv2d network.
S502, inputting the sunken point cloud image into the three-dimensional point cloud convolution network to obtain a second output result;
specifically, the three-dimensional point cloud convolution network is a PointConv network.
S503, inputting the first output result, the second output result and the depression extrusion force into the normalization layer for processing, and then inputting the normalization processing result into the vector splicing layer to obtain a splicing vector;
and S504, inputting the splicing vector into the full-connection layer to obtain a luggage soft and hard grade classification result of the luggage.
Preferably, the SoftMax function is arranged on the full link layer, the output result of the full link layer is mapped into the values (0,1) through the action of the SoftMax function, the sum of the values is 1 (the property of satisfying the probability), the sum can be understood as the probability, and when the output node is selected finally, the node with the maximum probability (namely the node with the maximum value corresponding to the probability) can be selected as the luggage soft and hard grade classification result of the luggage.
The classification recognition model is trained by training data which are manually marked, and is updated along with the expansion and the updating of the luggage database, and the classification recognition model is updated when the new increment of the luggage data in the luggage database exceeds a threshold value each time.
The above is an embodiment of the present application. The embodiments and specific parameters in the embodiments are only used for clearly illustrating the verification process of the application and are not used for limiting the patent protection scope of the application, which is defined by the claims, and all the equivalent structural changes made by using the contents of the specification and the drawings of the application should be included in the protection scope of the application.
Claims (10)
1. Civil aviation airport consignment luggage soft or hard package recognition device, its characterized in that includes:
a rack erected above the baggage conveying device;
the pressing and contacting module is arranged on the rack and used for contacting and extruding the luggage, so that the luggage generates concave deformation and the concave extrusion force at the concave position of the luggage is collected;
the distance measuring module is arranged on the rack and used for sensing the passing of the luggage and collecting the height distance between the upper surface of the luggage and the lower end of the pressure contact module;
the visual module is arranged on the rack and used for acquiring a concave RGB image and a concave point cloud image of the luggage extrusion concave position.
2. The civil aviation airport checked baggage hard-soft pack identification device of claim 1, wherein the press-touch module comprises:
the linear driving mechanism is vertically arranged on the rack;
and the contact rod is in driving connection with the movable end of the linear driving mechanism through a pressure sensor.
3. The civil aviation airport checked baggage hard and soft bag identification device of claim 2, wherein:
the linear driving mechanism is a servo electric cylinder.
4. The civil aviation airport checked baggage hard and soft baggage claim 2, wherein the press-touch module further comprises a press-touch control module, and the press-touch control module is configured to control a descending speed of the contact rod based on the height distance and the concave squeezing force, and specifically comprises:
if the height distance is greater than a preset safety distance, the descending speed is a first speed;
if the height distance is smaller than a preset safety distance, the descending speed is a second speed;
if the depression extrusion force is equal to the preset safety pressure, the descending speed is zero;
wherein the first speed is greater than the second speed.
5. The apparatus of claim 1, further comprising a baggage conveyer control module, wherein the baggage conveyer control module comprises:
a size recognition unit for calculating a baggage length of the baggage along a conveying direction through the baggage point cloud image acquired by the vision module;
a time control unit for controlling the time of the clock based on the function
Calculating the time interval
Wherein d represents the length of the luggage,
representing a conveying speed of the baggage conveying device;
and the starting and stopping control unit is used for suspending the luggage conveying equipment after the interval time when the ranging module senses that the luggage passes.
6. The hard and soft bag identification method of the consignment luggage in the civil aviation airport, which is based on the hard and soft bag identification device of the consignment luggage in the civil aviation airport of any one of claims 1 to 5, is characterized by comprising the following steps:
acquiring depression data of a luggage squeezing position, wherein the depression data comprises depression squeezing force, a depression RGB image and a depression point cloud image of the luggage squeezing depression;
comparing and screening the luggage data and sample data in a sample set based on a similarity algorithm, wherein the sample data comprises sample extrusion force, a sample RGB image, a sample point cloud image and a sample hardness grade;
if the sample data similar to the luggage data is screened out, taking the sample hardness level of the sample data as the luggage hardness level of the luggage;
and if sample data similar to the luggage data is not screened out, inputting the luggage data into a classification identification model to obtain the luggage soft and hard grade of the luggage.
7. The method of claim 6, wherein the step of inputting the baggage data into a classification recognition model to obtain the baggage soft and hard levels of the baggage further comprises:
and storing the baggage data and the corresponding baggage soft and hard grades into the sample set.
8. The method of claim 6, wherein the comparing and screening the baggage data with the sample data in the sample set based on the similarity algorithm comprises:
acquiring a difference value between the depression extrusion force and the sample extrusion force, and acquiring a first similarity based on the difference value;
inputting the concave RGB image and the sample RGB image into a Siamese network model to obtain a second similarity;
carrying out similarity calculation on the sunken point cloud image and the sample point cloud image to obtain a third similarity;
and comparing the first similarity, the second similarity and the third similarity with preset similar conditions to judge whether the luggage data is similar to the sample data.
9. The method for identifying soft and hard bags of baggage consigned by civil aviation airport according to claim 8, wherein the specific formula for calculating the similarity of the sunken point cloud image and the sample point cloud image is as follows:
wherein r (A, B) represents a third similarity, A represents a sunken point cloud image, B represents a sample point cloud image, and the higher the value of r (A, B), the more similar the sunken point cloud image A and the sample point cloud image B are;
the weight of the point is represented by a weight,
the distance of the point cloud is represented,
representing the ith sub-point cloud in the N sub-point cloud sets obtained by K-means clustering the concave point cloud image A,
the point of maximum depth in the vertical direction, i.e. the central position of the push rod,
representing a hyper-parameter;
wherein,
the concrete formula of (1) is as follows:
wherein h represents an adjustable parameter;
wherein,
the concrete formula of (1) is as follows:
wherein,
representing the distance between the point cloud X and the point cloud Y, X representing the points in the point cloud subset X, and Y representing the points in the point cloud subset Y.
10. The civil aviation airport checked bag soft and hard bag identification method of claim 6, wherein the classification identification model comprises a two-dimensional convolution network, a three-dimensional point cloud convolution network, a normalization layer, a vector splicing layer and a full connection layer; inputting the luggage data into a classification recognition model, and obtaining the luggage hardness and softness grade of the luggage comprises the following steps:
inputting the sunken RGB image into the two-dimensional convolution network to obtain a first output result;
inputting the sunken point cloud image into the three-dimensional point cloud convolution network to obtain a second output result;
inputting the first output result, the second output result and the depression extrusion force into the normalization layer for processing, and then inputting the normalization processing result into the vector splicing layer to obtain a splicing vector;
and inputting the splicing vector into the full-connection layer to obtain the luggage soft and hard grade classification result of the luggage.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211036891.7A CN115112508B (en) | 2022-08-29 | 2022-08-29 | Device and method for identifying soft and hard bags of consigned luggage in civil aviation airport |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211036891.7A CN115112508B (en) | 2022-08-29 | 2022-08-29 | Device and method for identifying soft and hard bags of consigned luggage in civil aviation airport |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115112508A true CN115112508A (en) | 2022-09-27 |
| CN115112508B CN115112508B (en) | 2023-01-24 |
Family
ID=83335523
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211036891.7A Active CN115112508B (en) | 2022-08-29 | 2022-08-29 | Device and method for identifying soft and hard bags of consigned luggage in civil aviation airport |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115112508B (en) |
Citations (34)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5306306A (en) * | 1991-02-13 | 1994-04-26 | Lunar Corporation | Method for periprosthetic bone mineral density measurement |
| JP2966832B1 (en) * | 1998-04-17 | 1999-10-25 | 株式会社パルタック | Method and apparatus for sorting packaged goods by store |
| US20020014533A1 (en) * | 1995-12-18 | 2002-02-07 | Xiaxun Zhu | Automated object dimensioning system employing contour tracing, vertice detection, and forner point detection and reduction methods on 2-d range data maps |
| JP2008262804A (en) * | 2007-04-12 | 2008-10-30 | Sony Corp | Battery pack |
| JP2008268075A (en) * | 2007-04-23 | 2008-11-06 | Toshiba Corp | Non-destructive inspection method, and non-destructive inspection device |
| CN101846503A (en) * | 2010-04-21 | 2010-09-29 | 中国科学院自动化研究所 | Luggage information on-line obtaining system based on stereoscopic vision and method thereof |
| WO2011011894A1 (en) * | 2009-07-31 | 2011-02-03 | Optosecurity Inc. | Method and system for identifying a liquid product in luggage or other receptacle |
| US20130048457A1 (en) * | 2011-08-25 | 2013-02-28 | Tammi Sbordoni | Flexible and Transparent Articles of Luggage |
| US20140002239A1 (en) * | 2012-06-27 | 2014-01-02 | Treefrog Developments, Inc. | Tracking and control of personal effects |
| GB2523633A (en) * | 2013-12-24 | 2015-09-02 | Vanguard Identification Systems Inc | Electronic luggage tags |
| US20160227897A1 (en) * | 2013-09-11 | 2016-08-11 | British Airways Plc | Identification apparatus and method |
| CN106056056A (en) * | 2016-05-23 | 2016-10-26 | 浙江大学 | Long-distance non-contact luggage volume detection system and method thereof |
| CN106696248A (en) * | 2015-11-13 | 2017-05-24 | 通用汽车环球科技运作有限责任公司 | Additive manufacturing of a body component on a tube frame |
| US10014076B1 (en) * | 2015-02-06 | 2018-07-03 | Brain Trust Innovations I, Llc | Baggage system, RFID chip, server and method for capturing baggage data |
| CN108991697A (en) * | 2018-07-27 | 2018-12-14 | 浙江师范大学 | It is a kind of that luggage case method is followed based on Gait Recognition automatically |
| CN109443670A (en) * | 2018-09-13 | 2019-03-08 | 肇庆学院 | A kind of body of a motor car surface indentation and dent test method |
| CN109506746A (en) * | 2018-11-02 | 2019-03-22 | 北京小米移动软件有限公司 | The control method and device of luggage case |
| CN109858437A (en) * | 2019-01-30 | 2019-06-07 | 苏州大学 | Based on the baggage volume automatic classification method for generating inquiry network |
| CN110785362A (en) * | 2017-06-30 | 2020-02-11 | 松下知识产权经营株式会社 | Projection indicating device, cargo sorting system and projection indicating method |
| US10592713B1 (en) * | 2018-12-31 | 2020-03-17 | Datalogic Usa, Inc. | Computationally-augmented video display facilitating package handling |
| CN211376032U (en) * | 2019-12-27 | 2020-08-28 | 上海航空印刷有限公司 | PET bottom luggage strip |
| CN111598063A (en) * | 2020-07-22 | 2020-08-28 | 北京纳兰德科技股份有限公司 | Luggage category determination method and device |
| US20200302589A1 (en) * | 2019-03-22 | 2020-09-24 | Idemia Identity & Security France | Baggage identification method |
| CN111783569A (en) * | 2020-06-17 | 2020-10-16 | 天津万维智造技术有限公司 | Luggage specification detection and personal bag information binding method of self-service consignment system |
| CN111899258A (en) * | 2020-08-20 | 2020-11-06 | 广东机场白云信息科技有限公司 | Self-service consignment luggage specification detection method |
| CN112565616A (en) * | 2021-03-01 | 2021-03-26 | 民航成都物流技术有限公司 | Target grabbing method, system and device and readable storage medium |
| CN112801050A (en) * | 2021-03-29 | 2021-05-14 | 民航成都物流技术有限公司 | Intelligent luggage tracking and monitoring method and system |
| CN113012136A (en) * | 2021-03-24 | 2021-06-22 | 中国民航大学 | Airport luggage counting method and counting system based on target detection |
| CN213544237U (en) * | 2020-09-17 | 2021-06-25 | 嘉兴市万事发包装制品有限公司 | Strength testing device of suitcase wheelset |
| CN113272682A (en) * | 2018-12-21 | 2021-08-17 | 莱卡地球系统公开股份有限公司 | Reality capture with laser scanner and camera |
| CN113602799A (en) * | 2021-08-05 | 2021-11-05 | 西南科技大学 | Airport luggage case carrying system and control method thereof |
| CN114170456A (en) * | 2021-10-22 | 2022-03-11 | 深圳先进技术研究院 | Hardness identification method, hardness identification system, hardness identification device and storage medium |
| CN114160447A (en) * | 2021-11-20 | 2022-03-11 | 中云智慧（北京）科技有限公司 | Advanced machine inspection system and method |
| CN114359876A (en) * | 2022-03-21 | 2022-04-15 | 成都奥伦达科技有限公司 | Vehicle target identification method and storage medium |
-
2022
- 2022-08-29 CN CN202211036891.7A patent/CN115112508B/en active Active
Patent Citations (34)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5306306A (en) * | 1991-02-13 | 1994-04-26 | Lunar Corporation | Method for periprosthetic bone mineral density measurement |
| US20020014533A1 (en) * | 1995-12-18 | 2002-02-07 | Xiaxun Zhu | Automated object dimensioning system employing contour tracing, vertice detection, and forner point detection and reduction methods on 2-d range data maps |
| JP2966832B1 (en) * | 1998-04-17 | 1999-10-25 | 株式会社パルタック | Method and apparatus for sorting packaged goods by store |
| JP2008262804A (en) * | 2007-04-12 | 2008-10-30 | Sony Corp | Battery pack |
| JP2008268075A (en) * | 2007-04-23 | 2008-11-06 | Toshiba Corp | Non-destructive inspection method, and non-destructive inspection device |
| WO2011011894A1 (en) * | 2009-07-31 | 2011-02-03 | Optosecurity Inc. | Method and system for identifying a liquid product in luggage or other receptacle |
| CN101846503A (en) * | 2010-04-21 | 2010-09-29 | 中国科学院自动化研究所 | Luggage information on-line obtaining system based on stereoscopic vision and method thereof |
| US20130048457A1 (en) * | 2011-08-25 | 2013-02-28 | Tammi Sbordoni | Flexible and Transparent Articles of Luggage |
| US20140002239A1 (en) * | 2012-06-27 | 2014-01-02 | Treefrog Developments, Inc. | Tracking and control of personal effects |
| US20160227897A1 (en) * | 2013-09-11 | 2016-08-11 | British Airways Plc | Identification apparatus and method |
| GB2523633A (en) * | 2013-12-24 | 2015-09-02 | Vanguard Identification Systems Inc | Electronic luggage tags |
| US10014076B1 (en) * | 2015-02-06 | 2018-07-03 | Brain Trust Innovations I, Llc | Baggage system, RFID chip, server and method for capturing baggage data |
| CN106696248A (en) * | 2015-11-13 | 2017-05-24 | 通用汽车环球科技运作有限责任公司 | Additive manufacturing of a body component on a tube frame |
| CN106056056A (en) * | 2016-05-23 | 2016-10-26 | 浙江大学 | Long-distance non-contact luggage volume detection system and method thereof |
| CN110785362A (en) * | 2017-06-30 | 2020-02-11 | 松下知识产权经营株式会社 | Projection indicating device, cargo sorting system and projection indicating method |
| CN108991697A (en) * | 2018-07-27 | 2018-12-14 | 浙江师范大学 | It is a kind of that luggage case method is followed based on Gait Recognition automatically |
| CN109443670A (en) * | 2018-09-13 | 2019-03-08 | 肇庆学院 | A kind of body of a motor car surface indentation and dent test method |
| CN109506746A (en) * | 2018-11-02 | 2019-03-22 | 北京小米移动软件有限公司 | The control method and device of luggage case |
| CN113272682A (en) * | 2018-12-21 | 2021-08-17 | 莱卡地球系统公开股份有限公司 | Reality capture with laser scanner and camera |
| US10592713B1 (en) * | 2018-12-31 | 2020-03-17 | Datalogic Usa, Inc. | Computationally-augmented video display facilitating package handling |
| CN109858437A (en) * | 2019-01-30 | 2019-06-07 | 苏州大学 | Based on the baggage volume automatic classification method for generating inquiry network |
| US20200302589A1 (en) * | 2019-03-22 | 2020-09-24 | Idemia Identity & Security France | Baggage identification method |
| CN211376032U (en) * | 2019-12-27 | 2020-08-28 | 上海航空印刷有限公司 | PET bottom luggage strip |
| CN111783569A (en) * | 2020-06-17 | 2020-10-16 | 天津万维智造技术有限公司 | Luggage specification detection and personal bag information binding method of self-service consignment system |
| CN111598063A (en) * | 2020-07-22 | 2020-08-28 | 北京纳兰德科技股份有限公司 | Luggage category determination method and device |
| CN111899258A (en) * | 2020-08-20 | 2020-11-06 | 广东机场白云信息科技有限公司 | Self-service consignment luggage specification detection method |
| CN213544237U (en) * | 2020-09-17 | 2021-06-25 | 嘉兴市万事发包装制品有限公司 | Strength testing device of suitcase wheelset |
| CN112565616A (en) * | 2021-03-01 | 2021-03-26 | 民航成都物流技术有限公司 | Target grabbing method, system and device and readable storage medium |
| CN113012136A (en) * | 2021-03-24 | 2021-06-22 | 中国民航大学 | Airport luggage counting method and counting system based on target detection |
| CN112801050A (en) * | 2021-03-29 | 2021-05-14 | 民航成都物流技术有限公司 | Intelligent luggage tracking and monitoring method and system |
| CN113602799A (en) * | 2021-08-05 | 2021-11-05 | 西南科技大学 | Airport luggage case carrying system and control method thereof |
| CN114170456A (en) * | 2021-10-22 | 2022-03-11 | 深圳先进技术研究院 | Hardness identification method, hardness identification system, hardness identification device and storage medium |
| CN114160447A (en) * | 2021-11-20 | 2022-03-11 | 中云智慧（北京）科技有限公司 | Advanced machine inspection system and method |
| CN114359876A (en) * | 2022-03-21 | 2022-04-15 | 成都奥伦达科技有限公司 | Vehicle target identification method and storage medium |
Non-Patent Citations (4)
| Title |
|---|
| NILS BOYSEN 等: "Automated sortation conveyors:A survey from an operational research perspective", 《EUROPEAN JOURNAL OF OPERATIONAL RESEARCH》 * |
| 朱嘉宸: "基于深度相机的机场自助行李托运关键技术的研究", 《中国优秀硕士学位论文全文数据库工程科技II辑》 * |
| 杜明谦 等: "RFID技术在机场行李自动分拣系统中的应用", 《电讯技术》 * |
| 甄军平等: "北京大兴国际机场托运行李自动识别技术方案浅析", 《智能建筑》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115112508B (en) | 2023-01-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110390691B (en) | Ore dimension measuring method based on deep learning and application system | |
| CN106548182B (en) | Pavement crack detection method and device based on deep learning and main cause analysis | |
| CN114266884A (en) | Method for detecting sorting target of multi-form bottle-shaped articles positioned by rotating frame | |
| CN110246122A (en) | Small size bearing quality determining method, apparatus and system based on machine vision | |
| CN112330607A (en) | Coal and gangue identification method, device and system based on image identification technology | |
| KR102391501B1 (en) | Classification System and method for atypical recycled goods using Deep learning | |
| CN113807466B (en) | Logistics package autonomous detection method based on deep learning | |
| CN112850186B (en) | Mixed pile-dismantling method based on 3D vision | |
| CN113688825A (en) | AI intelligent garbage recognition and classification system and method | |
| CN116665011A (en) | Coal flow foreign matter identification method for coal mine belt conveyor based on machine vision | |
| CN116129135A (en) | Tower crane safety early warning method based on small target visual identification and virtual entity mapping | |
| CN111597857B (en) | Logistics package detection method, device, equipment and readable storage medium | |
| CN111814739A (en) | Method, device and equipment for detecting express package volume and storage medium | |
| Moirogiorgou et al. | Intelligent robotic system for urban waste recycling | |
| CN115641555B (en) | Truck overload overrun early warning method based on AI vision | |
| CN115112508B (en) | Device and method for identifying soft and hard bags of consigned luggage in civil aviation airport | |
| CN116468703A (en) | Floor slab reinforcement acceptance method and system based on machine vision | |
| CN110143447A (en) | Zinc ingot metal pile offline transhipment reinforcement means automatically | |
| CN113554706B (en) | Trolley parcel position detection method based on deep learning | |
| CN117735244A (en) | Box-type cargo intelligent inspection integrated system and method | |
| CN116674091B (en) | Automatic aggregate feeding method for post-material yard of concrete mixing station | |
| CN113192058B (en) | Intelligent brick pile loading system based on computer vision and loading method thereof | |
| CN115953568A (en) | Intelligent mineral sorting method based on lightweight target detection framework | |
| CN116258908A (en) | Ground disaster prediction evaluation classification method based on unmanned aerial vehicle remote sensing image data | |
| Yang et al. | Design of an Airport Baggage Soft and Hard Bag Recognition Device |
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 |