CN112347985A - Material type detection method and device - Google Patents
Material type detection method and device Download PDFInfo
- Publication number
- CN112347985A CN112347985A CN202011370344.3A CN202011370344A CN112347985A CN 112347985 A CN112347985 A CN 112347985A CN 202011370344 A CN202011370344 A CN 202011370344A CN 112347985 A CN112347985 A CN 112347985A
- Authority
- CN
- China
- Prior art keywords
- image
- preset
- determining
- identified
- recognition
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 239000000463 material Substances 0.000 title claims abstract description 220
- 238000001514 detection method Methods 0.000 title claims description 9
- 238000012544 monitoring process Methods 0.000 claims abstract description 56
- 238000000034 method Methods 0.000 claims abstract description 41
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 32
- 238000012545 processing Methods 0.000 claims abstract description 12
- 230000015654 memory Effects 0.000 claims description 20
- 230000007246 mechanism Effects 0.000 claims description 6
- 238000010276 construction Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 8
- 238000012549 training Methods 0.000 description 8
- 230000008569 process Effects 0.000 description 7
- 239000004035 construction material Substances 0.000 description 4
- 239000013077 target material Substances 0.000 description 4
- 229910000831 Steel Inorganic materials 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 239000010959 steel Substances 0.000 description 3
- 238000009435 building construction Methods 0.000 description 2
- 230000008034 disappearance Effects 0.000 description 2
- 239000011521 glass Substances 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- 241000287828 Gallus gallus Species 0.000 description 1
- 229910001294 Reinforcing steel Inorganic materials 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000007635 classification algorithm Methods 0.000 description 1
- 238000013145 classification model Methods 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 238000005070 sampling 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
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
Landscapes
- Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a method and a device for detecting material types, wherein the method for detecting the material types comprises the following steps: acquiring a monitoring video of a preset monitoring area; determining a first image containing a material to be identified by adopting an image difference algorithm based on a monitoring video; and inputting the first image into a preset image recognition model for material recognition, and determining a material type recognition result of the material to be recognized. By implementing the method and the device, the automatic identification of the class of the material is realized, the workload of constructors is reduced, the material identification can be carried out on the image containing the material by automatically extracting the image in the monitoring video and processing the image, and the problem of identification errors caused by manual identification can be effectively avoided.
Description
Technical Field
The invention relates to the technical field of building construction, in particular to a material type detection method and device.
Background
On a construction site, it is often necessary to transport the material to a designated work surface for work. When the materials are transported to the operation surface, the type of the current construction materials of the operation surface can be judged only by means of observation of construction personnel through glasses, and then the current construction progress, material cost accounting statistics and the like can be deduced by utilizing the type of the current materials. However, because the construction site is often in high-altitude operation, and certain similarity exists in the appearances of different construction materials, when the materials are transported to the position near the operation surface, the identification difficulty of constructors is greatly increased, the mode of manually identifying the materials increases the workload of the constructors, and the identification accuracy is difficult to guarantee.
Disclosure of Invention
In view of this, the embodiment of the invention provides a method and a device for detecting material types, which solve the problems of large workload and low identification accuracy caused by manual material identification in the prior art.
According to a first aspect, an embodiment of the present invention provides a method for detecting a material category, including:
acquiring a monitoring video of a preset monitoring area;
determining a first image containing a material to be identified by adopting an image difference algorithm based on the monitoring video;
and inputting the first image into a preset image recognition model for material recognition, and determining a material type recognition result of the material to be recognized.
Optionally, the determining, based on the monitoring video, the first image containing the material to be identified by using an image difference algorithm includes:
acquiring two frames of images adjacent to a preset time interval in the monitoring video;
calculating the image difference degree between the two frames of images by adopting an image difference algorithm;
and when the image difference degree of the two frames of images is greater than a preset difference degree threshold value, determining that the next frame of image is the first image.
Optionally, the inputting the first image into a preset image recognition model for material recognition, and determining a material type recognition result of the material to be recognized includes:
acquiring a template image of a target putting area corresponding to the material to be identified;
performing template matching on the template image and the first image according to a template matching algorithm, and determining a second image corresponding to the material area to be identified;
and inputting the second image into a preset image recognition model for material recognition, and determining a material type recognition result of the material to be recognized.
Optionally, the template matching the template image and the first image according to a template matching algorithm to determine a second image corresponding to the material region to be identified includes:
determining a template matching area of the template image on the first image according to a template matching algorithm;
performing size expansion on the template matching area according to a preset expansion proportion;
and determining the second image according to the coordinate position of the template matching area after the size expansion and the first image.
Optionally, the preset expansion ratio is obtained by:
acquiring first size information of the target putting area and second size information of all types of materials;
and determining the preset expansion ratio according to the first size information and the second size information.
Optionally, the inputting the second image into a preset image recognition model for material recognition, and determining a material type recognition result of the material to be recognized includes:
acquiring a preset number of predicted material category identification results, wherein the predicted material category identification results are the predicted material category identification results output by inputting a preset number of second images into a preset image identification model;
and voting the predicted material category identification result according to a preset voting mechanism, and determining the material category identification result of the material to be identified.
Optionally, the preset image recognition model is a ResNet50 model.
According to a second aspect, an embodiment of the present invention provides a material type detection apparatus, including:
the first acquisition module is used for acquiring a monitoring video of a preset monitoring area of the first tower crane;
the first processing module is used for determining a first image containing the material to be identified by adopting an image difference algorithm based on the monitoring video;
and the second processing module is used for inputting the first image into a preset image recognition model for material recognition, and determining a material type recognition result of the material to be recognized.
According to a third aspect, embodiments of the present invention provide a non-transitory computer readable storage medium storing computer instructions which, when executed by a processor, implement the method of the first aspect of the present invention and any one of its alternatives.
According to a fourth aspect, an embodiment of the present invention provides an electronic device, including: a memory and a processor, the memory and the processor being communicatively coupled to each other, the memory having stored therein computer instructions, the processor being configured to execute the computer instructions to perform the method of the first aspect of the present invention and any one of the alternatives thereof.
The technical scheme of the invention has the following advantages:
the embodiment of the invention provides a material type detection method and device, wherein a monitoring video of a preset monitoring area is obtained, then a first image containing a material to be identified is determined by adopting an image difference algorithm based on the monitoring video, and then a preset image identification model is adopted to identify the first image, so that a material type identification result of the material to be identified is obtained. Therefore, automatic identification of the category to which the material belongs is achieved, workload of constructors is reduced, material identification can be carried out on the image containing the material by automatically extracting the image in the monitoring video and processing the image, and the problem of identification errors caused by manual identification can be effectively avoided.
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, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
FIG. 1 is a flow chart of a method for detecting material type according to an embodiment of the present invention;
FIG. 2 is a schematic structural diagram of a tower crane in the embodiment of the invention;
FIG. 3 is a first image extracted from a surveillance video according to an embodiment of the present invention;
FIG. 4 is an image of a template of a hook according to an embodiment of the present invention;
FIG. 5 is a second image corresponding to the material region to be identified after template matching in the embodiment of the present invention;
FIG. 6 is a schematic structural diagram of a material type detection apparatus according to an embodiment of the present disclosure;
fig. 7 is a schematic structural diagram of an electronic device in an embodiment of the present invention.
Detailed Description
The technical solutions of the present invention will be described clearly and completely with reference to the accompanying drawings, and it should be understood that the described embodiments are some, but not all embodiments of the present invention. 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.
In addition, the technical features involved in the different embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.
On a construction site, it is often necessary to transport the material to a designated work surface for work. When the materials are transported to the operation surface, the type of the current construction materials of the operation surface can be judged only by means of observation of construction personnel through glasses, and then the current construction progress, material cost accounting statistics and the like can be deduced by utilizing the type of the current materials. However, because the construction site is often in high-altitude operation, and certain similarity exists in the appearances of different construction materials, when the materials are transported to the position near the operation surface, the identification difficulty of constructors on the ground is greatly increased, the mode of manually identifying the materials increases the workload of the constructors, and the identification accuracy is difficult to ensure.
Based on the above problem, an embodiment of the present invention provides a method for detecting a material type, as shown in fig. 1, the method specifically includes the following steps:
step S101: and acquiring a monitoring video of a preset monitoring area. Specifically, the preset monitoring area may be disposed near a current working surface area of the building so as to monitor the working surface materials, and specifically, video monitoring may be performed through a camera installed near the working surface. The tower crane is structurally schematically illustrated in fig. 2, a lifting hook 11 and a material 12 to be identified, which is connected with the lifting hook 11, are arranged on a lifting arm 1 of the tower crane, and the material 12 to be identified is transported to an operation surface area through the lifting hook 11 to perform operation. The preset monitoring area may also be a monitoring range of a camera arranged on a boom of the tower crane apparatus shown in fig. 2, as long as monitoring of the material of the working surface transported by the hook can be achieved, and the invention is not limited thereto.
Step S102: and determining a first image containing the material to be identified by adopting an image difference algorithm based on the monitoring video. In the process of transporting materials of an operation surface by a tower crane, after the materials enter a preset monitoring area, a moving hook and the transported materials to be identified can appear in a monitoring picture, so that a video image containing the materials to be identified can be extracted from the monitoring video, an image difference algorithm is adopted to judge that when a moving object with a size larger than a certain area appears in a camera area, a video image intercepting operation is carried out to obtain a first image containing the materials to be identified, and the materials to be identified are borne by the hook, so that in practical application, the first image containing the materials to be identified can also be extracted from the video image with the hook by judging whether the hook appears in the monitoring picture, and the invention is not limited by the method.
Step S103: and inputting the first image into a preset image recognition model for material recognition, and determining a material type recognition result of the material to be recognized. Specifically, the preset image recognition model example may be selected from image classification algorithms in the prior art according to actual requirements, and the established image recognition model includes, for example: the ResNet50 algorithm, etc., and the present invention is not limited thereto.
By executing the steps, the method for detecting the material type, provided by the embodiment of the invention, obtains the monitoring video of the preset monitoring area, then determines the first image containing the material to be identified by adopting the image difference algorithm based on the monitoring video, and then identifies the first image by adopting the preset image identification model to obtain the material type identification result of the material to be identified. Therefore, automatic identification of the category to which the material belongs is achieved, workload of constructors is reduced, material identification can be carried out on the image containing the material by automatically extracting the image in the monitoring video and processing the image, and the problem of identification errors caused by manual identification can be effectively avoided.
Specifically, in an embodiment, the step S102 specifically includes the following steps:
step S201: two frames of images adjacent to a preset time interval in the monitoring video are obtained. Specifically, when the tower crane does not work or the lifting hook and the borne operation surface material do not enter a preset monitoring area, the picture of the image in the monitoring video is only an environment background image, after the material to be identified enters the monitoring video, the moving material appears in the monitoring video picture, therefore, whether the material is monitored or not can be judged by acquiring two frames of images at a certain time interval, the preset time interval can be set according to the actual rising speed of the lifting hook, and whether the lifting hook and the material appear or not can be determined through the two frames of images.
Step S202: and calculating the image difference degree between the two frames of images by adopting an image difference algorithm. Specifically, since the background image is fixed, the difference between the two frames of images only lies in the hook and the transported material, an image difference algorithm can be adopted to perform difference on corresponding pixel points on the two frames of images, and since the pixel values of the background images are the same, the image areas of the moving target, namely the hook and the material, can be obtained according to the region formed by the pixel points with the pixel difference value not being 0 after difference is performed, and the ratio of the image area to the video image area can be used for representing the image difference between the two frames of images, namely, the larger the ratio is, the larger the difference is, and the smaller the ratio is, otherwise, the difference is. In practical application, in order to obtain a complete image containing a material, the area ratio of the minimum image area occupied by the hook and the material completely appearing in the video image to the whole video image may be set as a preset difference threshold, if the calculated image difference is greater than the preset difference threshold, it is indicated that the hook and the material completely appear in the next frame of image, and the image is used as an original image for identifying the material, as shown in fig. 3.
Specifically, in an embodiment, the step S103 specifically includes the following steps:
step S301: and acquiring a template image of the target putting area corresponding to the material to be identified. Specifically, in the embodiment of the present invention, since the material is transported to the target throwing area by the hook, the hook can be used as a fixed reference object of the material, and thus, the hook template image is used as the template image of the target throwing area corresponding to the material to be identified. In practical applications, the template image may also be a template image corresponding to a reference object with a fixed working area, for example: the invention is not limited to images of buildings at work sites.
Step S302: and carrying out template matching on the template image and the first image according to a template matching algorithm, and determining a second image corresponding to the material area to be identified. Specifically, because the environment of the building construction site is complex, the material identification is directly performed on the first image, and the accuracy of the material type identification result is seriously influenced due to serious image background interference. Therefore, in order to further improve the accuracy of the subsequent material type identification result, the embodiment of the invention provides a method for positioning the material area in the image by using the position corresponding relation between the lifting hook of the tower crane and the hanging object, namely the material to be identified, so that the second image only containing the material area is extracted, and the influence of the environment background in the image is eliminated.
Step S303: and inputting the second image into a preset image recognition model for material recognition, and determining a material type recognition result of the material to be recognized. Specifically, the second image is the image only containing the material area, so that the influence of the environment background in the first image is eliminated, the chicken image is input into the preset image recognition model for material recognition, and the accuracy of material recognition is improved.
Specifically, in an embodiment, the step S302 specifically includes the following steps:
step S01: and determining a template matching area of the template image on the first image according to a template matching algorithm. Specifically, the shape and the color of the lifting hook on the tower crane are basically the same in the same engineering project. Therefore, according to the actual item, a fixed hook image including only a hook may be selected as a matching template, and the image feature of the hook in the hook image shown in fig. 4 may be used to match the hook included in the first image, so as to obtain a template matching region that is an image region of the hook in the first image.
Step S02: and carrying out size expansion on the template matching area according to a preset expansion proportion. Because the position relation between the lifting hook and the material is relatively fixed, the area where the material is located can be obtained by expanding the template matching area where the lifting hook is located according to a certain proportion, as shown in fig. 5.
Specifically, in the embodiment of the present invention, the first size information of the target delivery area and the second size information of all types of materials may be obtained; and determining a preset expansion ratio according to the first size information and the second size information. For example, assuming that the size information of the lifting hook is the size of the lifting hook, the material with the largest size among all the material types transported by the lifting hook is assumed to be a reinforced material, and the relative position between the lifting hook and the material is fixed, the preset expansion ratio can be obtained by reserving a certain ratio threshold according to the ratio relationship between the size of the reinforced material and the size of the lifting hook, assuming that the length of the lifting hook is 1 meter in width of 3 meters, and the length of the material transported with the largest size is 1 meter in width of 3 meters, the expansion ratio is that the length along the direction of the position where the material is located is expanded by 3 times and the width is expanded by 3 times by taking the area where the lifting hook is located as the center. It should be noted that the preset expansion ratio can also be adjusted according to the actual engineering requirements, and the present invention is not limited thereto.
Step S03: and determining a second image according to the coordinate position of the template matching area after the size expansion and the first image. Specifically, according to the coordinate position of the template matching area in the first image after the size expansion, a second image corresponding to the material area can be extracted from the first image. The second image is utilized to identify the material, so that the interference of the environmental background on the material category identification result can be effectively avoided, and the accuracy of the material category identification result is improved.
Specifically, in an embodiment, the step S303 includes the following steps:
step S001: and acquiring a preset number of predicted material category identification results, wherein the predicted material category identification results are the predicted material category identification results output by inputting a preset number of second images into a preset image identification model. Specifically, as the lifting hook and the material need a certain time to move in the monitoring video, a plurality of first images containing the material to be identified at different moments can be obtained through the steps, a plurality of second images are further determined, and then each second image is respectively input into a preset image identification model to obtain a predicted material type identification result corresponding to the image. In the embodiment of the invention, in the process of extracting the second image by template matching of the image extracted from the monitoring video, the phenomenon of image gradient disappearance exists, so the preset image identification model adopted in the embodiment of the invention is a ResNet50 model, because the ResNet50 algorithm adopts a residual connection mode to solve the problem of gradient disappearance to a certain extent, the ResNet50 algorithm has more excellent identification performance compared with other algorithms, and the algorithm has the characteristic of easy expansion, is favorable for being fused with subsequent further operation aiming at the material type identification result, and is easy for system integration.
Step S002: and voting the predicted material category identification result according to a preset voting mechanism, and determining the material category identification result of the material to be identified. Specifically, in order to further improve the accuracy of identification and avoid identification errors possibly caused by single identification, in the embodiment of the present invention, the material category identification result to be identified is voted by using a voting mechanism, so as to determine the material category identification result of the material to be identified according to the voting result, and the voting mechanism may be set according to actual requirements, for example: the predicted material category identification result with the predicted material category identification result larger than a certain threshold may be used as the final material category identification result, or the predicted material category identification result with the highest frequency of occurrence may be used as the final material category identification result according to a minority-obeying majority principle, for example: 5 pictures are collected in one lifting process, one picture is identified as a steel bar, 4 pictures are identified as non-steel bars, the result of final voting is the non-steel bar, and the invention is not limited by the method. Thus, the recognition error rate can be reduced by using the voting mechanism.
Specifically, the preset image recognition model may be obtained by training in the following manner:
and acquiring a historical material image and a material type identification result of the corresponding target material. Specifically, the historical material image is an image corresponding to a material area of the material transported on the tower crane in the monitoring video, namely the second image, the historical material image is provided with marking information, and the marking information comprises a material type identification result of the target material in the historical material image, such as 'reinforcing steel bar' and the like.
And inputting the historical material image into a preset image recognition model, and determining a predicted material category recognition result. The initial parameters in the preset image recognition model can be set manually according to experience and can also be set in a random mode.
And updating parameters in the preset image recognition model according to the material type recognition result and the predicted material type recognition result of the target material. The method comprises the steps of inputting historical material images into a preset image recognition model to obtain a predicted material category recognition result, and correcting parameters of the preset image recognition model by using errors between the predicted material category recognition result and a material category recognition result of a target material. Specifically, in the training process, the sampling weight of the misjudged sample can be increased after training for a certain number of times, and meanwhile, the adoption weight is also increased for the difficult sample with high recognition difficulty, for example: sample images with wrong model identification and sample images with high difficulty in manual judgment and identification are subjected to image operation such as size conversion of the images to enrich the number of samples, and the number of samples with wrong judgment samples and difficult samples is increased, so that the training effect of the model is improved, and the training time is shortened. In addition, in the model training process, the identification effect of the training model can be displayed by adopting a thermodynamic diagram, and if the thermodynamic diagram is distributed in a target area, namely a material area, and the weight is larger, the model is good; if the thermodynamic diagram is weighted in the non-target area, the model is not good, and parameters or data are required to be adjusted again for training until the classification model with the thermodynamic diagram satisfying is trained. After the ResNet50 model trained by the embodiment of the invention is verified by the thermodynamic diagram, the classification result meets the actual requirements of engineering projects, and the type of the material can be accurately identified.
By executing the steps, the method for detecting the material type provided by the embodiment of the invention obtains the monitoring video of the preset monitoring area, then determines the first image containing the material to be identified by adopting the image difference algorithm based on the monitoring video, then performs template matching on the template image and the first image by utilizing the template matching algorithm, further determines the second image corresponding to the area where the material to be identified is located, and then identifies the second image by adopting the preset image identification model to obtain the material type identification result of the material to be identified. Thereby realized the tower crane transportation operation face material belonged to the automatic identification of classification, reduced constructor work load to through drawing the image is automatic in the surveillance video, and handle the image and can obtain the image that only contains the material and carry out material identification, thereby avoided external interference image such as environment to carry out the influence of material identification, improved material identification's accuracy, and can effectively avoid because the wrong problem of discernment that artifical discernment caused.
An embodiment of the present invention further provides a device for detecting a material type, as shown in fig. 6, the device for detecting a material type specifically includes:
the first obtaining module 101 is configured to obtain a monitoring video of a preset monitoring area of the first tower crane. For details, refer to the related description of step S101 in the above method embodiment, and no further description is provided here.
The first processing module 102 is configured to determine, based on the surveillance video, a first image containing the material to be identified by using an image difference algorithm. For details, refer to the related description of step S102 in the above method embodiment, and no further description is provided here.
The second processing module 103 is configured to input the first image into a preset image recognition model for material recognition, and determine a material type recognition result of the material to be recognized. For details, refer to the related description of step S103 in the above method embodiment, and no further description is provided here.
Further functional descriptions of the modules are the same as those of the corresponding method embodiments, and are not repeated herein.
Through the cooperative cooperation of the above components, the material type detection device provided by the embodiment of the invention obtains the monitoring video of the preset monitoring area, then determines the first image containing the material to be identified by adopting the image difference algorithm based on the monitoring video, and then identifies the first image by adopting the preset image identification model to obtain the material type identification result of the material to be identified. Therefore, automatic identification of the category to which the material belongs is achieved, workload of constructors is reduced, material identification can be carried out on the image containing the material by automatically extracting the image in the monitoring video and processing the image, and the problem of identification errors caused by manual identification can be effectively avoided.
An embodiment of the present invention further provides an electronic device, as shown in fig. 7, the electronic device may include a processor 901 and a memory 902, where the processor 901 and the memory 902 may be connected by a bus or in another manner, and fig. 7 takes the connection by the bus as an example.
The memory 902, which is a non-transitory computer readable storage medium, may be used to store non-transitory software programs, non-transitory computer executable programs, and modules, such as program instructions/modules corresponding to the methods in the embodiments of the present invention. The processor 901 executes various functional applications and data processing of the processor, i.e., implements the above-described method, by executing non-transitory software programs, instructions, and modules stored in the memory 902.
The memory 902 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application program required for at least one function; the storage data area may store data created by the processor 901, and the like. Further, the memory 902 may include high speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, flash memory device, or other non-transitory solid state storage device. In some embodiments, the memory 902 may optionally include memory located remotely from the processor 901, which may be connected to the processor 901 via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
One or more modules are stored in the memory 902, which when executed by the processor 901 performs the methods described above.
The specific details of the electronic device may be understood by referring to the corresponding related descriptions and effects in the above method embodiments, and are not described herein again.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware related to instructions of a computer program, and the program can be stored in a computer readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic Disk, an optical Disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a Flash Memory (Flash Memory), a Hard Disk (Hard Disk Drive, abbreviated as HDD) or a Solid State Drive (SSD), etc.; the storage medium may also comprise a combination of memories of the kind described above.
The above embodiments are only for illustrating the technical solutions of the present invention and not for limiting the same, and although the present invention is described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications and equivalents may be made to the embodiments of the invention without departing from the spirit and scope of the invention, which is to be covered by the claims.
Claims (10)
1. A material class detection method is characterized by comprising the following steps:
acquiring a monitoring video of a preset monitoring area;
determining a first image containing a material to be identified by adopting an image difference algorithm based on the monitoring video;
and inputting the first image into a preset image recognition model for material recognition, and determining a material type recognition result of the material to be recognized.
2. The method of claim 1, wherein determining the first image containing the material to be identified based on the surveillance video using an image differencing algorithm comprises:
acquiring two frames of images adjacent to a preset time interval in the monitoring video;
calculating the image difference degree between the two frames of images by adopting an image difference algorithm;
and when the image difference degree of the two frames of images is greater than a preset difference degree threshold value, determining that the next frame of image is the first image.
3. The method according to claim 1, wherein the inputting the first image into a preset image recognition model for material recognition, and determining a material type recognition result of the material to be recognized comprises:
acquiring a template image of a target putting area corresponding to the material to be identified;
performing template matching on the template image and the first image according to a template matching algorithm, and determining a second image corresponding to the material area to be identified;
and inputting the second image into a preset image recognition model for material recognition, and determining a material type recognition result of the material to be recognized.
4. The method according to claim 3, wherein the template matching the template image and the first image according to a template matching algorithm to determine a second image corresponding to the material region to be identified comprises:
determining a template matching area of the template image on the first image according to a template matching algorithm;
performing size expansion on the template matching area according to a preset expansion proportion;
and determining the second image according to the coordinate position of the template matching area after the size expansion and the first image.
5. The method of claim 4, wherein the predetermined expansion ratio is obtained by:
acquiring first size information of the target putting area and second size information of all types of materials;
and determining the preset expansion ratio according to the first size information and the second size information.
6. The method according to claim 3, wherein the inputting the second image into a preset image recognition model for material recognition and determining a material type recognition result of the material to be recognized comprises:
acquiring a preset number of predicted material category identification results, wherein the predicted material category identification results are the predicted material category identification results output by inputting a preset number of second images into a preset image identification model;
and voting the predicted material category identification result according to a preset voting mechanism, and determining the material category identification result of the material to be identified.
7. The method of claim 1, wherein the predetermined image recognition model is the ResNet50 model.
8. A material type detection device, comprising:
the first acquisition module is used for acquiring a monitoring video of a preset monitoring area of the first tower crane;
the first processing module is used for determining a first image containing the material to be identified by adopting an image difference algorithm based on the monitoring video;
and the second processing module is used for inputting the first image into a preset image recognition model for material recognition, and determining a material type recognition result of the material to be recognized.
9. A non-transitory computer-readable storage medium storing computer instructions that, when executed by a processor, implement the method of any one of claims 1-7.
10. An electronic device, comprising:
a memory and a processor, the memory and the processor being communicatively coupled to each other, the memory having stored therein computer instructions, the processor performing the method of any of claims 1-7 by executing the computer instructions.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011370344.3A CN112347985A (en) | 2020-11-30 | 2020-11-30 | Material type detection method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011370344.3A CN112347985A (en) | 2020-11-30 | 2020-11-30 | Material type detection method and device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112347985A true CN112347985A (en) | 2021-02-09 |
Family
ID=74365199
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011370344.3A Pending CN112347985A (en) | 2020-11-30 | 2020-11-30 | Material type detection method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112347985A (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113111921A (en) * | 2021-03-19 | 2021-07-13 | 中建科技集团有限公司 | Object recognition method, object recognition device, electronic equipment and storage medium |
CN113911916A (en) * | 2021-09-13 | 2022-01-11 | 杭州大杰智能传动科技有限公司 | Hook material combination positioning control method and system for intelligent tower crane |
CN114604787A (en) * | 2022-01-24 | 2022-06-10 | 杭州大杰智能传动科技有限公司 | Automatic material characteristic identification method and device for unmanned intelligent tower crane |
CN114604757A (en) * | 2022-01-24 | 2022-06-10 | 杭州大杰智能传动科技有限公司 | Magnetic-guiding intelligent tower crane hook suitable for steel structure materials and guiding method thereof |
CN115588121A (en) * | 2022-11-03 | 2023-01-10 | 腾晖科技建筑智能(深圳)有限公司 | Tower crane lifting object type detection method and system based on sensing data and image sequence |
CN117726882A (en) * | 2024-02-07 | 2024-03-19 | 杭州宇泛智能科技有限公司 | Tower crane object identification method, system and electronic equipment |
WO2024066664A1 (en) * | 2022-09-26 | 2024-04-04 | 三一汽车制造有限公司 | Concrete pumpability category identification method and apparatus, and electronic device |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2002253539A (en) * | 2001-03-05 | 2002-09-10 | Nippon Telegr & Teleph Corp <Ntt> | Medical image identification system, medical image identification processing method, program for medical image identification, and recording medium therefor |
CN102521565A (en) * | 2011-11-23 | 2012-06-27 | 浙江晨鹰科技有限公司 | Garment identification method and system for low-resolution video |
CN109241985A (en) * | 2017-07-11 | 2019-01-18 | 普天信息技术有限公司 | A kind of image-recognizing method and device |
CN109740517A (en) * | 2018-12-29 | 2019-05-10 | 上海依图网络科技有限公司 | A kind of method and device of determining object to be identified |
CN111126396A (en) * | 2019-12-25 | 2020-05-08 | 北京科技大学 | Image recognition method and device, computer equipment and storage medium |
CN111444806A (en) * | 2020-03-19 | 2020-07-24 | 成都云盯科技有限公司 | Commodity touch information clustering method, device and equipment based on monitoring video |
CN111476064A (en) * | 2019-01-23 | 2020-07-31 | 北京奇虎科技有限公司 | Small target detection method and device, computer equipment and storage medium |
CN111598091A (en) * | 2020-05-20 | 2020-08-28 | 北京字节跳动网络技术有限公司 | Image recognition method and device, electronic equipment and computer readable storage medium |
-
2020
- 2020-11-30 CN CN202011370344.3A patent/CN112347985A/en active Pending
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2002253539A (en) * | 2001-03-05 | 2002-09-10 | Nippon Telegr & Teleph Corp <Ntt> | Medical image identification system, medical image identification processing method, program for medical image identification, and recording medium therefor |
CN102521565A (en) * | 2011-11-23 | 2012-06-27 | 浙江晨鹰科技有限公司 | Garment identification method and system for low-resolution video |
CN109241985A (en) * | 2017-07-11 | 2019-01-18 | 普天信息技术有限公司 | A kind of image-recognizing method and device |
CN109740517A (en) * | 2018-12-29 | 2019-05-10 | 上海依图网络科技有限公司 | A kind of method and device of determining object to be identified |
CN111476064A (en) * | 2019-01-23 | 2020-07-31 | 北京奇虎科技有限公司 | Small target detection method and device, computer equipment and storage medium |
CN111126396A (en) * | 2019-12-25 | 2020-05-08 | 北京科技大学 | Image recognition method and device, computer equipment and storage medium |
CN111444806A (en) * | 2020-03-19 | 2020-07-24 | 成都云盯科技有限公司 | Commodity touch information clustering method, device and equipment based on monitoring video |
CN111598091A (en) * | 2020-05-20 | 2020-08-28 | 北京字节跳动网络技术有限公司 | Image recognition method and device, electronic equipment and computer readable storage medium |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113111921A (en) * | 2021-03-19 | 2021-07-13 | 中建科技集团有限公司 | Object recognition method, object recognition device, electronic equipment and storage medium |
CN113111921B (en) * | 2021-03-19 | 2024-06-14 | 中建科技集团有限公司 | Object identification method, device, electronic equipment and storage medium |
CN113911916A (en) * | 2021-09-13 | 2022-01-11 | 杭州大杰智能传动科技有限公司 | Hook material combination positioning control method and system for intelligent tower crane |
CN113911916B (en) * | 2021-09-13 | 2023-06-02 | 杭州大杰智能传动科技有限公司 | Lifting hook material combined positioning control method and system for intelligent tower crane |
CN114604787A (en) * | 2022-01-24 | 2022-06-10 | 杭州大杰智能传动科技有限公司 | Automatic material characteristic identification method and device for unmanned intelligent tower crane |
CN114604757A (en) * | 2022-01-24 | 2022-06-10 | 杭州大杰智能传动科技有限公司 | Magnetic-guiding intelligent tower crane hook suitable for steel structure materials and guiding method thereof |
CN114604787B (en) * | 2022-01-24 | 2023-06-02 | 杭州大杰智能传动科技有限公司 | Material automatic characteristic identification method and device for unmanned intelligent tower crane |
CN114604757B (en) * | 2022-01-24 | 2023-06-02 | 杭州大杰智能传动科技有限公司 | Magnetic guiding intelligent tower crane lifting hook suitable for steel structure materials and guiding method thereof |
WO2024066664A1 (en) * | 2022-09-26 | 2024-04-04 | 三一汽车制造有限公司 | Concrete pumpability category identification method and apparatus, and electronic device |
CN115588121A (en) * | 2022-11-03 | 2023-01-10 | 腾晖科技建筑智能(深圳)有限公司 | Tower crane lifting object type detection method and system based on sensing data and image sequence |
CN115588121B (en) * | 2022-11-03 | 2023-07-04 | 腾晖科技建筑智能(深圳)有限公司 | Tower crane object type detection method and system based on sensing data and image sequence |
CN117726882A (en) * | 2024-02-07 | 2024-03-19 | 杭州宇泛智能科技有限公司 | Tower crane object identification method, system and electronic equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112347985A (en) | Material type detection method and device | |
Yang et al. | Deep concrete inspection using unmanned aerial vehicle towards cssc database | |
EP3937128A2 (en) | Image defect detection method and apparatus, electronic device, storage medium and product | |
CN111178206B (en) | Building embedded part detection method and system based on improved YOLO | |
US11783588B2 (en) | Method for acquiring traffic state, relevant apparatus, roadside device and cloud control platform | |
CN114926726B (en) | Unmanned ship sensing method based on multitask network and related equipment | |
CN114994061A (en) | Machine vision-based steel rail intelligent detection method and system | |
CN116052097A (en) | Map element detection method and device, electronic equipment and storage medium | |
CN116630367B (en) | Target tracking method, device, electronic equipment and storage medium | |
CN113256683A (en) | Target tracking method and related equipment | |
CN112507956A (en) | Signal lamp identification method and device, electronic equipment, road side equipment and cloud control platform | |
CN116486290A (en) | Unmanned aerial vehicle monitoring and tracking method and device, electronic equipment and storage medium | |
CN115019252B (en) | Concrete quality detection method and device and monitoring equipment | |
EP3930306A2 (en) | Method and apparatus for stabilizing image, roadside device and cloud control platform | |
CN111860084A (en) | Image feature matching and positioning method and device and positioning system | |
CN111985497B (en) | Crane operation identification method and system under overhead transmission line | |
CN116543327A (en) | Method, device, computer equipment and storage medium for identifying work types of operators | |
CN112308061B (en) | License plate character recognition method and device | |
TW202213251A (en) | Reinforced frame automatic inspection system, computer readable storage device and operation method thereof | |
CN110705479A (en) | Model training method, target recognition method, device, equipment and medium | |
CN114581890B (en) | Method and device for determining lane line, electronic equipment and storage medium | |
CN115994928B (en) | Target tracking method, device, equipment and medium | |
CN117975343B (en) | Monitoring data processing method and system for building construction | |
CN117274843B (en) | Unmanned aerial vehicle front end defect identification method and system based on lightweight edge calculation | |
CN115187957A (en) | Ground element detection method, device, equipment, medium and product |
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 |