CN117457032A - Storage medium destroying method based on volume identification - Google Patents
Storage medium destroying method based on volume identification Download PDFInfo
- Publication number
- CN117457032A CN117457032A CN202311785374.4A CN202311785374A CN117457032A CN 117457032 A CN117457032 A CN 117457032A CN 202311785374 A CN202311785374 A CN 202311785374A CN 117457032 A CN117457032 A CN 117457032A
- Authority
- CN
- China
- Prior art keywords
- storage medium
- destroying
- volume identification
- identification system
- destruction
- 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 50
- 230000006378 damage Effects 0.000 claims abstract description 52
- 230000005540 biological transmission Effects 0.000 claims abstract description 27
- 238000011156 evaluation Methods 0.000 claims abstract description 10
- 239000012634 fragment Substances 0.000 claims description 67
- 239000013598 vector Substances 0.000 claims description 30
- 238000010298 pulverizing process Methods 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 4
- 238000012545 processing Methods 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 24
- 230000008569 process Effects 0.000 description 15
- 238000013527 convolutional neural network Methods 0.000 description 11
- 238000011176 pooling Methods 0.000 description 9
- 230000004913 activation Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 5
- 230000003287 optical effect Effects 0.000 description 4
- 238000013135 deep learning Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000011524 similarity measure Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001502 supplementing effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11B—INFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORD CARRIER AND TRANSDUCER
- G11B5/00—Recording by magnetisation or demagnetisation of a record carrier; Reproducing by magnetic means; Record carriers therefor
- G11B5/02—Recording, reproducing, or erasing methods; Read, write or erase circuits therefor
- G11B5/024—Erasing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0475—Generative networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/094—Adversarial learning
-
- 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
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
-
- 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
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/761—Proximity, similarity or dissimilarity measures
-
- 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
- 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
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Medical Informatics (AREA)
- Multimedia (AREA)
- Databases & Information Systems (AREA)
- Image Analysis (AREA)
Abstract
The application discloses a storage medium destroying method based on volume identification, which relates to the technical field of image data processing; the method for destroying the storage medium based on the volume identification comprises the following steps: the storage medium to be destroyed enters destruction equipment from a transmission system, the storage medium to be destroyed on the transmission system is subjected to volume identification through a first volume identification system, and the storage medium to be destroyed is evaluated; when the destroying program is carried out, the second volume identification system arranged in the destroying equipment is used for carrying out volume identification on the storage medium in destroying, and the crushing granularity of the storage medium in the destroying equipment is evaluated based on the identification result of the second volume identification system; and judging whether the destruction of the storage medium meets the destruction requirement or not based on the identification result and the evaluation result of the second volume identification system.
Description
Technical Field
The application relates to the technical field of image data processing, in particular to a storage medium destroying method based on volume identification.
Background
The volume recognition technology is a technology for recognizing and classifying three-dimensional objects by using a deep learning algorithm, and can capture information of shapes, sizes, positions, directions and the like of the objects from multiple angles and distances to generate a volume (voxel) representation, namely a three-dimensional pixel grid. The volume identification technology can be applied to the storage medium destroying process, and the storage medium to be destroyed is evaluated and monitored, so that the thoroughly and safety of the destroying process are ensured.
In the prior art, the application research of the point cloud identification technology in the process of destroying the storage medium is less, and in the technical scheme recorded in the patent document with the publication number of CN114528950A, a storage medium destroying and identifying network model comprising a point cloud characteristic extracting network and a medium volume identifying network is adopted to identify the volume of the storage medium fragments in the mechanical crushing process, so that whether the volume of the medium fragments meets the destroying requirement or the secondary crushing is needed is automatically judged. In the technical scheme, the focus of body volume identification is to destroy fragments, and the focus monitoring object of point cloud identification is a storage medium in the destruction process, namely the destruction fragments.
However, in the process of destroying the storage media, the destroying requirements, namely the destroying fragment size, of different storage media are different; in addition, for batch destroying tasks simultaneously containing multiple types and specifications of storage media, identification and distinction are needed, so that a proper destroying mode and a proper destroying standard are adopted, and when the storage media are destroyed, corresponding destroying programs and destroying requirements still need to be manually selected.
Disclosure of Invention
The technical scheme mainly provides a storage medium destroying method based on volume identification, which automatically selects corresponding destroying standards based on the types and the quantity of the storage medium to be destroyed, which are identified by a first volume identification system; in the destroying process, whether the preset destroying standard is met or not is judged based on fragments of the storage medium identified by the second volume identification system, and human intervention on a destroying program is reduced.
In order to achieve the above purpose, the present application provides the following technical solutions:
a method of destroying a storage medium based on volume identification, comprising:
the method comprises the steps that a storage medium to be destroyed enters destruction equipment from a transmission system, the storage medium to be destroyed on the transmission system is subjected to volume identification through a first volume identification system arranged on one side of the transmission system so as to obtain an image or a feature vector of the storage medium, the storage medium to be destroyed is evaluated based on an identification result, and the evaluation result comprises the category and the number of the storage medium; wherein the destruction device is capable of pulverizing the storage medium into fragments;
selecting a corresponding destruction requirement based on the identification result and the evaluation result of the first body identification system;
when the destroying program is carried out, carrying out volume identification on the destroyed storage medium based on a second volume identification system arranged in the destroying equipment so as to acquire images or feature vectors of fragments of the storage medium, and evaluating the crushing granularity of the storage medium in the destroying equipment based on the identification result of the second volume identification system;
and judging whether the destruction of the storage medium meets the destruction requirement or not based on the identification result and the evaluation result of the second volume identification system, and matching the identified storage medium fragments with the storage medium identified in the first volume identification system by the second volume identification system.
Preferably, the matching method comprises the following steps:
training a generation countermeasure network by taking the storage medium image or the feature vector identified by the first body identification system as input so that the generation countermeasure network can generate an image or the feature vector of fragments corresponding to the storage medium;
taking the generated image or the feature vector as reference data of a second volume identification system;
when the second volume identification system identifies the patch data, a discriminator network is used to compare the patch data with the reference data to achieve an automatic match.
Preferably, when the first integral identification system identifies that the types of the storage media to be destroyed on the transmission system are multiple, the selection of the destruction program can be applicable to the multiple types of the storage media to be destroyed; and the selection of the destruction requirements can be applied to the various destruction media to be stored.
Preferably, the transmission system includes a control unit for controlling a transmission speed of the transmission system, the control unit being responsive to the first body recognition system.
Preferably, the storage area corresponding to the storage medium is judged based on the first volume identification system, and the storage area is coated with a mark recognizable by the second volume identification system.
Preferably, when the second volume identification system identifies that the number of fragments of the storage medium having a granularity greater than the first preset value is less than the second preset value; and when the duty ratio of the marked storage medium fragments in the storage medium fragments larger than the first preset value is smaller than a third preset value, the destroying procedure is completed.
Preferably, when the second volume identification system identifies that the number of fragments of the storage medium having a granularity greater than the first preset value is less than the second preset value; and when the number of the marked storage medium fragments with the granularity larger than the fourth preset value is smaller than the fifth preset value, the destroying program is completed.
Preferably, the first body recognition system adopts a multi-mode sensor, including a laser radar, a camera and an ultrasonic sensor.
Compared with the known public technology, the technical scheme provided by the application has the following beneficial effects: the storage medium destroying method based on volume identification can automatically select corresponding destroying requirements based on the types and the quantity of the storage mediums to be destroyed, which are identified by the first volume identification system, and destroying equipment destroys the storage mediums based on the destroying requirements; in the destroying process, whether the destroying requirement is met or not is judged based on fragments of the storage medium identified by the second volume identification system, so that human intervention on a destroying program is reduced; and for batch destruction containing different storage media, the selection of the destruction requirements can simultaneously meet the destruction standards of the plurality of storage media, and the intelligent identification of the mixed media can be realized, so that the destruction efficiency and quality are improved, and meanwhile, the personnel investment is reduced.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below. It is apparent that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained from these drawings without inventive effort for a person of ordinary skill in the art.
Fig. 1 is a schematic structural diagram of a storage medium destruction method based on volume identification according to an embodiment of the present application.
Fig. 2 is a destruction flow chart of a method for destroying a storage medium based on volume identification according to an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present application more clear, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. It will be apparent that the embodiments described are some, but not all, of the embodiments of the present application. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments herein without making any inventive effort, are intended to be within the scope of the present application.
The embodiment of the application describes a storage medium destroying method based on volume identification, which is mainly aimed at destroying storage media such as a U disk, a hard disk, an optical disk and an IC card, and is mainly suitable for batch destroying tasks simultaneously containing various types and specifications in the storage media.
The method mainly comprises the following steps:
s10: the storage medium to be destroyed enters the destroying device from the transmission system 100, the storage medium to be destroyed on the transmission system is subjected to volume recognition through a first volume recognition system 200 arranged at one side of the transmission system to obtain an image or a feature vector of the storage medium, the storage medium to be destroyed is evaluated based on a recognition result, and the evaluation result comprises the category and the number of the storage medium;
s20: selecting a destruction requirement corresponding to the destruction device 300 based on the identification result and the evaluation result of the first body identification system; wherein the destruction device is capable of pulverizing the storage medium into fragments;
s30: when the destroying program is performed, the second volume identification system 400 arranged in the destroying device is used for carrying out volume identification on the destroyed storage medium so as to acquire images or feature vectors of fragments of the storage medium, and the crushing granularity of the storage medium in the destroying device is evaluated based on the identification result of the second volume identification system;
s40: and judging whether the destruction of the storage medium meets the destruction requirement or not based on the identification result and the evaluation result of the second volume identification system.
In the above technical solution, the first body identification system is used for classifying and counting the storage media to be destroyed, so as to select a proper destruction program and parameters. In the embodiment of the application, the first body recognition system adopts an image recognition algorithm based on deep learning, images or feature vectors of storage media on a transmission system are captured, then the images are subjected to feature extraction and classification through a convolutional neural network, and finally the types and the quantity of the storage media are output.
In the embodiment of the application, the first body identification system adopts a multi-mode sensor, including a laser radar, a camera and an ultrasonic sensor.
The identifying step of the first body identification system to the storage medium specifically includes the following aspects:
s11: preparing a data set containing different kinds of storage medium images and corresponding tags, for example, 0 represents a hard disk, 1 represents an optical disk, 2 represents a USB flash disk, and 3 represents an IC card; wherein the labels 0,1,2,3 are used to indicate the type of storage medium, i.e. each image has a corresponding label indicating which storage medium it belongs to. The purpose of this is to enable the convolutional neural network to learn from images and labels, thereby enabling classification and identification of storage media.
S12: a convolutional neural network structure, such as selecting the number, size, parameters, etc. of convolutional layers, pooling layers, full-connection layers, etc., is designed, and this structure can refer to the existing convolutional neural network model, such as AlexNet, VGGNet, resNet, etc.
S13: the convolutional neural network is trained by the data set, namely, the output of the network is calculated through forward propagation, and then the parameters of the network are updated through backward propagation, so that the network can learn the characteristics in the data set, and the classification accuracy is improved.
S14: the storage medium image on the transmission system is captured and then preprocessed, e.g., cropped, scaled, denoised, normalized, etc., so that the image meets the input requirements of the network. And then inputting the images into a trained convolutional neural network to obtain the output of the network, namely the type and the number of the storage media.
The convolutional neural network mainly performs multi-layer nonlinear transformation on the image, so that key features in the image are extracted. The convolutional neural network consists of a plurality of convolutional layers, a pooling layer, an activation layer and a full connection layer, wherein each layer has a plurality of learnable parameters such as convolutional kernels, offsets, weights and the like.
The function of the convolution layer is to perform local linear combination on the image, so as to extract low-level features such as edges, textures, shapes and the like of the image, wherein the output of the convolution layer can be expressed as:
；
wherein,represent the firstkThe convolution kernel is atiLayer numberjOutput of column, +.>Represent the firstkInput image of individual channels,>represent the firstkThe convolution kernel is atpFirst of the channelsmLine 1nWeight of column->Represent the firstkThe offset of the individual convolution kernels is such that,MandNrepresenting the height and width of the convolution kernel,Prepresenting the number of channels of the input image.
The effect of the pooling layer is to downsample the output of the convolution layer, thereby reducing the amount of computation and the number of parameters, increasing the robustness and perceptibility of the model, where the output of the pooling layer can be expressed as:
；
wherein,represent the firstkThe first channel is atiLine 1jOutput of column, +.>Represent the firstkThe input image of the individual channels is displayed,MandNrepresenting the height and width of the pooling nucleus,Sthe pooling step size is represented, and the maximum pooling is exemplified in the above formula.
The function of the activation layer is to perform nonlinear transformation on the output of the convolution layer or the pooling layer, thereby increasing the expressive power and complexity of the model. Typical activation functions include ReLU, sigmoid, tanh, and as an example, the ReLU function, the output of the activation layer is:
；
wherein,represent the firstkThe first channel is atiLine 1jOutput of column, +.>Represent the firstkThe input image of the individual channels is displayed,findicating activation of work function.
The function of the fully connected layer is to flatten the output of the convolution layer or pooling layer into a one-dimensional vector, and then to perform linear combination, so as to extract the advanced features of the image, and the output of the fully connected layer can be expressed as:
；
wherein,represent the firstiThe value of the output node,/>Represent the firstjValues of the input nodes +.>Represent the firstjInput nodes to the firstiWeights of the individual output nodes, +.>Represent the firstiThe offset of the output nodes of the circuit,Jrepresenting the number of input nodes; information on the type, number, location, etc. of the storage medium can thus be given.
To train convolutional neural networks, a loss function is defined to measure the difference between the predicted result and the real label. The loss function employed in the embodiments of the present application is expressed as:
；
wherein,Lthe value of the loss function is represented,represent the firstiThe true tag of the individual category,>represent the firstiThe probability of prediction for the individual categories,Kthe number of categories is indicated.
The function of the optimizer is to update the parameters of the convolutional neural network according to the loss function so that the loss function reaches the minimum value, and the optimizer adopted in the embodiment of the application is expressed as:
；
wherein,represent the firsttParameter values after +1 iterations, +.>Represent the firsttParameter values after several iterations,/->Indicates learning rate (I/O)>Representing the gradient of the loss function versus the parameter.
In the actual destroying process, different destroying standards exist for storage media with different types and different storage contents, for example, in BMB21-2007, how much different types of storage media need to be crushed when crushed by adopting a crushing method is given; in addition, different enterprises have different requirements on the crushing degree, and the area of each surface of the limited fragments is generally less than or equal to Xmm 2 For example, in the embodiment of the present application, for several storage media, the media should be crushed to an area of not more than 2mm when the usb disk is destroyed 2 When the hard disk is destroyed, the medium is crushed until the area of each surface is not more than 1mm 2 When the optical disk is destroyed, the medium is crushed to each surface area not more than 5mm 2 When the IC card is destroyed, the medium is crushed until the area of each surface is not more than 0.5mm 2 . Under the above limitation, in step S20 of the embodiment of the present application, when the destruction requirement is selected based on the identification result of the first integral identification system, for example, if the identification result is that the storage medium to be destroyed is a hard disk, the destruction requirement corresponding to the hard disk is selected, that is, the storage medium needs to be crushed until the area of each surface is not more than 1mm 2 The signal is transmitted to the destroying device, and the destroying device destroys the storage medium based on the destroying requirement.
In some batch destroying processes, simultaneous destroying of multiple storage mediums may be involved, so that when the first integral identifying system identifies that the types of the storage mediums to be destroyed on the transmission system are multiple, the selection of destroying requirements can be applicable to the multiple storage destroying mediums. That is, taking the above-mentioned destruction requirement as an example, when the first integral identification system identifies that the U disk, the hard disk and the optical disk exist on the transmission system, the medium should be crushed to an area of not more than 1mm 2 。
In this embodiment of the present application, the second volumetric identification system mainly evaluates the granularity of the storage medium to determine whether the storage medium fragments meet the destruction requirement, so that the identification method mainly adopted by the second volumetric identification system is based on pattern matching of images or feature vectors, that is, by comparing the images or feature vectors of the storage medium fragments with preset standard images or feature vectors, to determine whether the sizes and shapes of the storage medium fragments meet the destruction requirement.
In order to reduce the calculation amount of the second volume identification system, in the embodiment of the application, when the size of the crushed storage medium in the destructing device is evaluated in step S30, the second volume identification system matches the identified partial storage medium fragments with the storage medium identified in the first volume identification system.
Specifically, the matching method comprises the following steps:
s31: a generation model is trained, which is capable of generating an image or feature vector of a patch corresponding to the storage medium, using the storage medium image or feature vector identified by the first body recognition system as input.
S32: taking the fragment data generated by the generation model and the fragment data identified by the second volume identification system as inputs, training a judgment model so that the judgment model can judge whether the given fragment data is from the generation model or the second volume identification system;
s33: the generating model and the judging model form an objective function, and the generating model and the judging model are alternately optimized and trained, so that the generating model can generate more realistic fragment data, and the judging model can more accurately distinguish the sources of the fragment data. When the objective function reaches an equilibrium state, the distribution of the fragment data generated by the generated model and the fragment data identified by the second volume identification system should be identical, so that the purpose of automatic matching is achieved.
That is, in a conventional generation countermeasure network, the loss function of the generation model G is:
；
wherein,zis a random noise vector which is used to generate a random noise vector,is the distribution of noise->Is the output of the generation model G,is the probability of judgment of the output of the generated model G by the judgment model D.
The loss function of the discriminant function D is:
；
wherein,xis a real fragment data of the fragment,is the distribution of real data, +.>Is the judgment probability of the judgment model D on the real data, < >>Is the output of the generative model G, +.>Is the probability of judgment of the output of the generated model G by the judgment model D.
The goal of the generator is to generate a random noise vectorzIn generating dummy dataAnd makes the arbiter unable to distinguish between true and false, i.e. minimize +.>. The aim of the arbiter is to accurately judge the input dataxWhether true or generated and gives a probability +.>I.e. maximize +.>And->This objective function can be expressed as:
wherein,Eindicating the desire.
The significance of the objective function is that when the generator and the arbiter reach Nash equilibrium, the distribution of the generated data and the distribution of the real data are consistent, so that the purpose of data generation is realized.
In an embodiment of the present application, in order to achieve automatic matching using a generation countermeasure network, it is necessary to take an image or a feature vector of a storage medium recognized by a first volume recognition system as an input of a generator G, instead of a random noise vectorzThe method comprises the steps of carrying out a first treatment on the surface of the In this way, the information of the storage medium can be used to guide the generator G to generate a more realistic fragment image or feature vector. Its objective function can be expressed as:
；
wherein,yis an image or feature vector of the storage medium as perceived by the first volumetric recognition system,xis a picture or feature vector of the tile,is an image or feature vector of the generated fragment, < >>Is the probability value given by the arbiter.
In order to compare the patch data with the reference data using a network of discriminators or a similarity measure function, it is necessary to use the image or feature vector of the patch identified by the second volume identification system as input to the discriminator D, to compare with the image or feature vector of the patch generated by the generator G, and if the probability value given by the discriminator D is close to 1, this means that the matching degree between the two is high, otherwise, the matching degree is low. Of course, other similarity measure functions, such as euclidean distance, cosine similarity, etc. may be used to calculate the difference between the two. If the difference is small, the matching degree of the two is high, otherwise, the matching degree is low.
According to the technical scheme, the fact that the second volume identification system identifies fragments and the first volume identification system identifies images can be achieved, training of the convolutional neural network is not needed to be conducted on the fragments identified by the second volume identification system, and calculated amount of the second volume identification system is reduced. On the other hand, the destruction effect can be verified through the matching degree of the fragments identified by the second volume identification system and the pictures identified by the first volume identification system, and whether the expected destruction target is reached can be verified from another angle. In a further more important aspect, for a part of storage media containing more important or sensitive information, after the destruction is completed, a compliance check is required, so that fragments are identified by the second volume identification system to be recombined with a part of the picture identified by the first volume identification system, and the function of supplementing the destroyed compliance can be also achieved.
In some embodiments, the transport system comprises a control unit capable of controlling the transport speed of the transport system, the control unit being responsive to a selection of a destruction requirement. In the actual destroying process, the transmission system and the destroying device are generally started at the same time, when the destroying requirement corresponding to the storage medium to be destroyed is higher, the transmission system is often required to transmit at a slower speed, and the control unit is arranged to enable the destroying system to automatically select proper destroying requirements according to the type and the number of the storage medium to be destroyed, and also automatically adjust the transmission speed of the transmission device according to the destroying requirements, so that the manual intervention on the destroying system is further reduced.
Based on the foregoing explanation, a storage mediumIn the destroying process, different destroying standards are correspondingly different based on different types of storage media and different storage contents, for example, when the hard disk is limited to be destroyed, the media should be crushed to each area not exceeding 1mm 2 . It will be appreciated that this requirement is not limited to all hard disk fragments having an area of no more than 1mm 2 It is generally required that the area of more than a certain proportion, such as 90% of the hard disk fragments in the destruction device is not more than 1mm 2 And (3) obtaining the product.
In addition, for a specific storage medium, such as a hard disk, not all areas are important areas for pulverization, and it is always desirable that the pulverization of the data storage area can meet a preset requirement in the pulverization process.
Thus, in an embodiment of the present application, a storage area of a corresponding storage medium is determined based on a first volume identification system, and the storage area is coated with a marking that is identifiable by a second volume identification system.
The first volume identification system can judge the storage area of the storage medium by comparing the information of the type, the model and the like of the storage medium with the pre-stored basic information of the storage medium, and in addition, it can be understood that the application of the mark cannot influence the second volume identification system to identify and judge fragments of the storage medium, and cannot influence the fragment data identified by the second volume identification system to be matched with the image of the storage medium identified by the first volume identification system.
When judging whether the destruction of the storage medium meets the destruction requirement or not, the judging method provided by one embodiment of the application is that when the second volume identification system identifies that the number of the storage medium fragments with the granularity larger than a first preset value is smaller than a second preset value; and when the duty ratio of the marked storage medium fragments in the storage medium fragments larger than the first preset value is smaller than a third preset value, the destroying procedure is completed.
In the above technical solution, the setting of the first preset value, the second preset value, and the third preset value is set in advance according to different storage media.
Still use the hard disk as the storage medium, destroy the area of the identifiable surface that should smash the medium to 90% in the fragment when the requirement is destroying is not more than 1mm 2 This requirement is exemplified by the first preset value selecting the area of the identifiable surface to be 1mm 2 The second preset value is 10%, and the third preset value is 30%; the method for judging whether the destruction of the storage medium meets the destruction requirement should be: the second volume identification system identifies an area greater than 1mm 2 The number of storage medium fragments of less than 10%, wherein the identifiable surface areas are greater than 1mm 2 The fraction of marked storage medium fragments in the storage medium fragments is less than 30% (i.e. the fraction is less than 3% of all fragments).
The discrimination requirement can focus the focus of attention on the storage area of the storage medium, and is more suitable for the practical purpose of destroying the storage medium.
Based on the above principle, a discriminating method according to another embodiment of the present application is that when the second volume identification system identifies that the number of fragments of the storage medium having a granularity greater than the first preset value is smaller than the second preset value; and when the number of the marked storage medium fragments with the granularity larger than the fourth preset value is smaller than the fifth preset value, the destroying program is completed.
Continuing to use the hard disk as the storage medium, and destroying the storage medium with the requirement that the area of 90% of identifiable surfaces in fragments is not more than 1mm when the storage medium is destroyed 2 This requirement is exemplified by the first preset value selecting the area of the identifiable surface to be 1mm 2 The second preset value is 10%, and the fourth preset value is 0.5mm 2 The fifth preset value is 10%, then this criterion is actually: the second volume identification system identifies an area greater than 1mm 2 The number of storage medium fragments is less than 10%; and the second volume identification system identifies, among the marked pieces of storage medium, an area greater than 0.5mm 2 The number of storage media is less than 10%.
In this technical scheme, all the crushed fragment standards are distinguished from the crushed fragment standards of the storage area, and the destroying program needs to meet the two standards simultaneously to reach the preset destroying requirement.
By combining the above, the storage medium destroying method based on volume identification according to the embodiment of the present application can automatically select a corresponding destroying requirement based on the type and the number of the storage medium to be destroyed identified by the first volume identification system, and the destroying device destroys the storage medium based on the destroying requirement; in the destroying process, whether the destroying requirement is met or not is judged based on fragments of the storage medium identified by the second volume identification system, so that human intervention on a destroying program is reduced; in addition, for batch destruction containing different storage media, the destruction requirements can be selected to meet the destruction standards of the plurality of storage media at the same time, so that the intelligent identification of the mixed media can be realized, the destruction efficiency and quality are improved, and the personnel investment is reduced; in addition, through the second volume identification system, the fragments of the storage medium are matched with the original storage medium, so that traceability and verifiability of the destruction process are realized.
Moreover, in the various embodiments described above, relational terms such as first, second, and the like may be used solely to distinguish one operation or element or module from another operation or element or module without necessarily requiring or implying any actual such relationship or order between such elements or modules or operations.
The above embodiments are only for illustrating the technical solution of the present application, and are not limiting thereof; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; these modifications or substitutions do not depart from the essence of the corresponding technical solutions from the protection scope of the technical solutions of the embodiments of the present application.
Claims (8)
1. A method for destroying a storage medium based on volume identification, comprising:
the method comprises the steps that a storage medium to be destroyed enters destruction equipment from a transmission system, the storage medium to be destroyed on the transmission system is subjected to volume identification through a first volume identification system arranged on one side of the transmission system so as to obtain an image or a feature vector of the storage medium, the storage medium to be destroyed is evaluated based on an identification result, and the evaluation result comprises the category and the number of the storage medium; wherein the destruction device is capable of pulverizing the storage medium into fragments;
selecting a corresponding destruction requirement based on the identification result and the evaluation result of the first body identification system;
when the destroying program is carried out, carrying out volume identification on the destroyed storage medium based on a second volume identification system arranged in the destroying equipment so as to acquire images or feature vectors of fragments of the storage medium, and evaluating the crushing granularity of the storage medium in the destroying equipment based on the identification result of the second volume identification system;
and judging whether the destruction of the storage medium meets the destruction requirement or not based on the identification result and the evaluation result of the second volume identification system, and matching the identified storage medium fragments with the storage medium identified in the first volume identification system by the second volume identification system.
2. The method for destroying storage media based on volume identification according to claim 1, wherein said method for matching comprises:
training a generation countermeasure network by taking the storage medium image or the feature vector identified by the first body identification system as input so that the generation countermeasure network can generate an image or the feature vector of fragments corresponding to the storage medium;
taking the generated image or the feature vector as reference data of a second volume identification system;
when the second volume identification system identifies the patch data, a discriminator network is used to compare the patch data with the reference data to achieve an automatic match.
3. The method for destroying storage media based on volume identification according to claim 1, wherein when the first volume identification system identifies that there are a plurality of types of storage media to be destroyed on the transmission system, the selection of the destruction program can be applied to the plurality of types of storage media to be destroyed; and the selection of the destruction requirements can be applied to the various destruction media to be stored.
4. The method for destroying storage media based on volume identification according to claim 1, wherein said transmission system comprises a control unit for controlling a transmission speed of the transmission system, said control unit being responsive to said first volume identification system.
5. The method for destroying storage media based on volume identification according to claim 1, wherein the storage area of the corresponding storage media is judged based on the first volume identification system, and a mark recognizable by the second volume identification system is coated on the storage area.
6. The method of claim 5, wherein when the second volume identification system identifies that the number of storage medium fragments having a granularity greater than a first preset value is less than a second preset value; and when the duty ratio of the marked storage medium fragments in the storage medium fragments larger than the first preset value is smaller than a third preset value, the destroying procedure is completed.
7. The method of claim 5, wherein when the second volume identification system identifies that the number of storage medium fragments having a granularity greater than a first preset value is less than a second preset value; and when the number of the marked storage medium fragments with the granularity larger than the fourth preset value is smaller than the fifth preset value, the destroying program is completed.
8. The method for destroying storage media based on volumetric identification according to claim 1, wherein said first volumetric identification system employs a multi-modal sensor including lidar, cameras and ultrasonic sensors.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311785374.4A CN117457032B (en) | 2023-12-25 | 2023-12-25 | Storage medium destroying method based on volume identification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311785374.4A CN117457032B (en) | 2023-12-25 | 2023-12-25 | Storage medium destroying method based on volume identification |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117457032A true CN117457032A (en) | 2024-01-26 |
| CN117457032B CN117457032B (en) | 2024-03-22 |
Family
ID=89580317
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311785374.4A Active CN117457032B (en) | 2023-12-25 | 2023-12-25 | Storage medium destroying method based on volume identification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117457032B (en) |
Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080088873A1 (en) * | 2006-10-13 | 2008-04-17 | Fuji Xerox Co., Ltd. | Document administration system and document destruction apparatus |
| US8959182B1 (en) * | 2008-04-15 | 2015-02-17 | Crimson Corporation | Systems and methods for computer data recovery and destruction |
| CN105117302A (en) * | 2015-07-24 | 2015-12-02 | 厦门宇高信息科技有限公司 | Verification method for data destruction of storage medium |
| CN110196691A (en) * | 2019-06-11 | 2019-09-03 | 苏州浪潮智能科技有限公司 | The method and device of data destroying based on data level of confidentiality and hard disk media type |
| CN112287303A (en) * | 2020-12-24 | 2021-01-29 | 北京中超伟业信息安全技术股份有限公司 | Destroying equipment suitable for traceable confidential medium |
| US20210272589A1 (en) * | 2020-02-28 | 2021-09-02 | Sepehr Rajaie | Degausser Mechanism Utilizing Magnetic Fluid for Destroying the Functionality of Data Bearing Components of Magnetic Media Data Bearing Devices |
| CN113414136A (en) * | 2021-08-25 | 2021-09-21 | 阿里云计算有限公司 | Storage medium destroying method, device and system and storage medium |
| CN114153690A (en) * | 2021-11-18 | 2022-03-08 | 腾讯数码（天津）有限公司 | Program memory monitoring method and device, computer equipment and storage medium |
| CN114265669A (en) * | 2021-12-30 | 2022-04-01 | 胜斗士(上海)科技技术发展有限公司 | Destroyed container instance identification method and device |
| CN114528950A (en) * | 2022-04-24 | 2022-05-24 | 北京中超伟业信息安全技术股份有限公司 | Destroying method and system for identifying type of confidential medium based on three-dimensional point cloud |
| CN115512211A (en) * | 2022-08-24 | 2022-12-23 | 艾斯特国际安全技术（深圳）有限公司 | Certificate destruction method, system, device and storage medium |
| CN116071787A (en) * | 2023-01-06 | 2023-05-05 | 南京航空航天大学 | Multispectral palmprint recognition method, multispectral palmprint recognition system, electronic equipment and multispectral palmprint recognition medium |
| CN116561068A (en) * | 2023-06-30 | 2023-08-08 | 山东万里红信息技术有限公司 | Digital management method for automatically destroying medium |
| CN116702069A (en) * | 2023-08-08 | 2023-09-05 | 北京中超伟业信息安全技术股份有限公司 | Medium destroying method based on state sensing |
| WO2023173546A1 (en) * | 2022-03-15 | 2023-09-21 | 平安科技（深圳）有限公司 | Method and apparatus for training text recognition model, and computer device and storage medium |
| CN117235820A (en) * | 2023-09-26 | 2023-12-15 | 湖南惟储信息技术有限公司 | Chip self-destruction circuit and destruction method based on same |
-
2023
- 2023-12-25 CN CN202311785374.4A patent/CN117457032B/en active Active
Patent Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080088873A1 (en) * | 2006-10-13 | 2008-04-17 | Fuji Xerox Co., Ltd. | Document administration system and document destruction apparatus |
| US8959182B1 (en) * | 2008-04-15 | 2015-02-17 | Crimson Corporation | Systems and methods for computer data recovery and destruction |
| CN105117302A (en) * | 2015-07-24 | 2015-12-02 | 厦门宇高信息科技有限公司 | Verification method for data destruction of storage medium |
| CN110196691A (en) * | 2019-06-11 | 2019-09-03 | 苏州浪潮智能科技有限公司 | The method and device of data destroying based on data level of confidentiality and hard disk media type |
| US20210272589A1 (en) * | 2020-02-28 | 2021-09-02 | Sepehr Rajaie | Degausser Mechanism Utilizing Magnetic Fluid for Destroying the Functionality of Data Bearing Components of Magnetic Media Data Bearing Devices |
| CN112287303A (en) * | 2020-12-24 | 2021-01-29 | 北京中超伟业信息安全技术股份有限公司 | Destroying equipment suitable for traceable confidential medium |
| CN113414136A (en) * | 2021-08-25 | 2021-09-21 | 阿里云计算有限公司 | Storage medium destroying method, device and system and storage medium |
| CN114153690A (en) * | 2021-11-18 | 2022-03-08 | 腾讯数码（天津）有限公司 | Program memory monitoring method and device, computer equipment and storage medium |
| CN114265669A (en) * | 2021-12-30 | 2022-04-01 | 胜斗士(上海)科技技术发展有限公司 | Destroyed container instance identification method and device |
| WO2023173546A1 (en) * | 2022-03-15 | 2023-09-21 | 平安科技（深圳）有限公司 | Method and apparatus for training text recognition model, and computer device and storage medium |
| CN114528950A (en) * | 2022-04-24 | 2022-05-24 | 北京中超伟业信息安全技术股份有限公司 | Destroying method and system for identifying type of confidential medium based on three-dimensional point cloud |
| CN115512211A (en) * | 2022-08-24 | 2022-12-23 | 艾斯特国际安全技术（深圳）有限公司 | Certificate destruction method, system, device and storage medium |
| CN116071787A (en) * | 2023-01-06 | 2023-05-05 | 南京航空航天大学 | Multispectral palmprint recognition method, multispectral palmprint recognition system, electronic equipment and multispectral palmprint recognition medium |
| CN116561068A (en) * | 2023-06-30 | 2023-08-08 | 山东万里红信息技术有限公司 | Digital management method for automatically destroying medium |
| CN116702069A (en) * | 2023-08-08 | 2023-09-05 | 北京中超伟业信息安全技术股份有限公司 | Medium destroying method based on state sensing |
| CN117235820A (en) * | 2023-09-26 | 2023-12-15 | 湖南惟储信息技术有限公司 | Chip self-destruction circuit and destruction method based on same |
Non-Patent Citations (1)
| Title |
|---|
| 张茜: "数字档案馆风险管理研究", 博士电子期刊, 15 March 2022 (2022-03-15) * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117457032B (en) | 2024-03-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Gou et al. | Vehicle license plate recognition based on extremal regions and restricted Boltzmann machines | |
| CN101510257B (en) | Human face similarity degree matching method and device | |
| CN105719188B (en) | The anti-method cheated of settlement of insurance claim and server are realized based on plurality of pictures uniformity | |
| Anagnostopoulos et al. | A license plate-recognition algorithm for intelligent transportation system applications | |
| CN109711366B (en) | Pedestrian re-identification method based on group information loss function | |
| CN106408037B (en) | Image recognition method and device | |
| CN107679504A (en) | Face identification method, device, equipment and storage medium based on camera scene | |
| Shamrat et al. | A deep learning approach for face detection using max pooling | |
| CN113449704B (en) | Face recognition model training method and device, electronic equipment and storage medium | |
| CN109389105B (en) | Multitask-based iris detection and visual angle classification method | |
| CN111507320A (en) | Detection method, device, equipment and storage medium for kitchen violation behaviors | |
| CN111709404B (en) | Machine room legacy identification method, system and equipment | |
| Gu et al. | Embedded and real-time vehicle detection system for challenging on-road scenes | |
| CN114626476A (en) | Bird fine-grained image recognition method and device based on Transformer and component feature fusion | |
| CN117457032B (en) | Storage medium destroying method based on volume identification | |
| CN114724246A (en) | Dangerous behavior identification method and device | |
| CN116882486B (en) | Method, device and equipment for constructing migration learning weight | |
| CN117235820A (en) | Chip self-destruction circuit and destruction method based on same | |
| Rayavel et al. | Real time machine learning approach for a smart door unlocking using face recognition system | |
| CN113505716B (en) | Training method of vein recognition model, and recognition method and device of vein image | |
| Ali et al. | Image classification with Deep Convolutional Neural Network Using Tensorflow and Transfer of Learning | |
| CN110717544B (en) | Pedestrian attribute analysis method and system under vertical fisheye lens | |
| Chen et al. | Social distance monitoring of site workers for COVID-19 using context-guided data augmentation, deep learning, and homography transformation | |
| Abraham | Digital image forgery detection approaches: A review and analysis | |
| CN112818945A (en) | Convolutional network construction method suitable for subway station crowd counting |
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 |