CN110781812A - Method for automatically identifying target object by security check instrument based on machine learning - Google Patents
Method for automatically identifying target object by security check instrument based on machine learning Download PDFInfo
- Publication number
- CN110781812A CN110781812A CN201911017535.9A CN201911017535A CN110781812A CN 110781812 A CN110781812 A CN 110781812A CN 201911017535 A CN201911017535 A CN 201911017535A CN 110781812 A CN110781812 A CN 110781812A
- Authority
- CN
- China
- Prior art keywords
- image
- security check
- machine learning
- classifier
- target object
- 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
- 238000000034 method Methods 0.000 title claims abstract description 39
- 238000010801 machine learning Methods 0.000 title claims abstract description 33
- 238000000605 extraction Methods 0.000 claims abstract description 18
- 238000007689 inspection Methods 0.000 claims abstract description 5
- 238000012549 training Methods 0.000 claims description 42
- 238000004590 computer program Methods 0.000 claims description 19
- 238000012360 testing method Methods 0.000 claims description 19
- 230000006870 function Effects 0.000 claims description 9
- 238000002790 cross-validation Methods 0.000 claims description 6
- 238000011028 process validation Methods 0.000 claims description 2
- 238000004422 calculation algorithm Methods 0.000 abstract description 7
- 238000012545 processing Methods 0.000 abstract description 6
- 238000002591 computed tomography Methods 0.000 abstract description 2
- 239000013598 vector Substances 0.000 description 7
- 238000010586 diagram Methods 0.000 description 3
- 239000000243 solution Substances 0.000 description 3
- 239000012267 brine Substances 0.000 description 2
- 239000004927 clay Substances 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- HPALAKNZSZLMCH-UHFFFAOYSA-M sodium;chloride;hydrate Chemical compound O.[Na+].[Cl-] HPALAKNZSZLMCH-UHFFFAOYSA-M 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- OYPRJOBELJOOCE-UHFFFAOYSA-N Calcium Chemical compound [Ca] OYPRJOBELJOOCE-UHFFFAOYSA-N 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000000694 effects Effects 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
- 238000005457 optimization Methods 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 238000006467 substitution reaction 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
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N23/00—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00
- G01N23/02—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00 by transmitting the radiation through the material
- G01N23/04—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00 by transmitting the radiation through the material and forming images of the material
- G01N23/046—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00 by transmitting the radiation through the material and forming images of the material using tomography, e.g. computed tomography [CT]
-
- G01V5/226—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
- G06F18/24147—Distances to closest patterns, e.g. nearest neighbour classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/50—Extraction of image or video features by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
Abstract
The application provides a method for automatically identifying a target object by a security check instrument based on machine learning, which comprises the following steps: acquiring a CT image of an article to be detected, which is acquired by a security inspection instrument, and performing feature extraction on the CT image to obtain features of the CT image; identifying the characteristics of the CT image by using a preset classifier model to generate an identification result corresponding to the article to be detected; and sending the identification result to a security check instrument so that the security check instrument displays the identification result. The method applies machine learning to the field of CT scanning, fuses CT image processing and a machine learning algorithm, identifies a target object by using different classifiers through extracting the gray level characteristics and the directional gradient histogram characteristics of a CT image, and determines an optimal classifier model through comparing the recall rate and the accuracy rate; the classifier model determined by the method has higher accuracy in target identification.
Description
Technical Field
The application relates to the field of machine learning, in particular to a method for automatically identifying a target object by a security check instrument based on machine learning.
Background
CT scanners are now playing an increasingly important role in security systems. Compared with the traditional X-ray scanner, the CT scanner has the advantages of being more sensitive, more comprehensive and faster. Further, in combination with machine learning techniques, the recognition rate of CT scanners will be continuously improved. There are two main steps for an automatic object identification system of a CT scanner. The first is CT image feature extraction, and each CT image usually contains hundreds or thousands of slice images, unlike X-ray images, which makes the whole feature extraction process more complicated. Another step is the selection of the classifier, since the suitability of the classifier selection will directly affect the performance of the training. Therefore, we need to select the appropriate feature extraction algorithm and classifier to achieve the desired recall and accuracy.
Disclosure of Invention
The application aims to provide a method for automatically identifying a target object by a security check instrument based on machine learning, so that ideal recall rate and accuracy can be achieved in the process of automatically identifying the target object by security check.
In order to realize the task, the following technical scheme is adopted in the application:
in a first aspect, the present application provides a method for automatically identifying a target object by a security check instrument based on machine learning, including:
acquiring a CT image of an article to be detected, which is acquired by a security inspection instrument, and performing feature extraction on the CT image to obtain features of the CT image;
identifying the characteristics of the CT image by using a preset classifier model to generate an identification result corresponding to the article to be detected;
and sending the identification result to a security check instrument so that the security check instrument displays the identification result.
Further, before the identifying the features of the CT image by using the preset classifier model, the method further includes:
the features of the CT image are normalized to convert all feature values to values between 0 and 1.
Further, the training process of the preset classifier model includes:
acquiring CT images of a plurality of known articles by using a security check instrument, and extracting characteristics of each CT image to obtain the characteristics of the CT images;
establishing a data set, wherein the data set comprises a plurality of samples, and each sample comprises a characteristic corresponding to a known article and a label;
normalizing the characteristics in the data set, and then randomly dividing the data set into a training set and a testing set;
training different classifiers by using a training set to obtain classifier models, and testing the corresponding classifier models by using a test set;
and determining a classifier model from classifier models obtained by training different classifiers as the preset classifier model according to the recall rate and the accuracy.
Further, the feature extraction is performed on the CT image, wherein the extracted features are a gray scale feature and a histogram of oriented gradients feature of the CT image.
Further, the different classifiers include an SVM classifier and a KNN classifier.
Further, when the classifier is an SVM classifier, a Gaussian kernel function is adopted for training in the training process, and cross validation is performed on a training set by ten folds.
Further, when the classifier is a KNN classifier, selecting 5 as the value of K; and (5) taking the Euclidean distance as a parameter of the KNN classifier, and performing ten-fold cross validation on the training set.
In a second aspect, the present application provides an apparatus for automatically identifying a target object by a security check machine based on machine learning, including:
the acquisition module is used for acquiring a CT image of the to-be-detected article acquired by the security check instrument and extracting the characteristics of the CT image to obtain the characteristics of the CT image;
the recognition module is used for recognizing the characteristics of the CT image by utilizing a preset classifier model and generating a recognition result corresponding to the article to be detected;
and the display module is used for sending the identification result to a security check instrument so that the security check instrument displays the identification result.
Further, the apparatus further comprises a classifier model determination module comprising:
the acquisition and extraction module is used for acquiring CT images of a plurality of known articles by using the security check instrument and extracting characteristics of each CT image to obtain the characteristics of the CT image;
a data set establishing module for establishing a data set, wherein the data set comprises a plurality of samples, and each sample comprises a characteristic corresponding to a known article and a label;
the data set dividing module is used for carrying out normalization processing on the features in the data set and then randomly dividing the data set into a training set and a testing set;
the training and testing module is used for training different classifiers by utilizing a training set to obtain classifier models and testing the corresponding classifier models through a testing set;
and the classifier determining module is used for determining one classifier model from classifier models obtained by training different classifiers as the preset classifier model according to the recall rate and the accuracy rate.
In a third aspect, the present application provides a terminal device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the steps of the method for automatically identifying a target object by a machine learning-based security check instrument according to the first aspect when executing the computer program.
In a fourth aspect, the present application provides a computer-readable storage medium storing a computer program, which when executed by a processor implements the steps of the method for automatically identifying a target object by a machine learning-based security check instrument of the first aspect.
The application has the following technical characteristics:
the method applies machine learning to the field of CT scanning, fuses CT image processing and a machine learning algorithm, identifies a target object by using different classifiers through extracting the gray level characteristics and the directional gradient histogram characteristics of a CT image, and determines an optimal classifier model through comparing the recall rate and the accuracy rate; the classifier model determined by the method has higher accuracy in target identification.
Drawings
Fig. 1 is a schematic flowchart of a method for automatically identifying a target object by a security check machine based on machine learning according to an embodiment of the present disclosure;
FIG. 2 is a schematic diagram of a feature matrix for HOG feature extraction according to an embodiment of the present application;
FIG. 3 is a diagram illustrating support vectors according to an embodiment of the present application;
fig. 4 is a schematic flowchart of an apparatus for automatically identifying a target object by a security check machine based on machine learning according to an embodiment of the present application;
fig. 5 is a schematic structural diagram of a terminal device according to an embodiment of the present application.
Detailed Description
Referring to fig. 1, an embodiment of the present application provides a method for a security check instrument to automatically identify a target object based on machine learning, including:
s101, acquiring a CT image of an article to be detected, which is acquired by a security inspection instrument, and performing feature extraction on the CT image to obtain features of the CT image;
s102, recognizing the characteristics of the CT image by using a preset classifier model, and generating a recognition result corresponding to the article to be detected;
s103, sending the identification result to a security check instrument so that the security check instrument displays the identification result.
When the article to be detected passes through the security check instrument, an X-ray machine in the security check instrument is used for scanning to obtain a CT image of the article to be detected; before the CT image is identified, feature extraction is carried out on the CT image.
For different images, image features are one of its basic attributes, and image feature extraction is a process of converting features into numerical data. There are generally three basic features of an image: the gray scale feature, the texture feature and the shape feature are selected from the first two in the embodiment of the application.
The first method comprises the following steps: a grayscale feature. The gray feature extraction is a classic digital image processing algorithm, and because the feature only contains the statistical feature of gray distribution, the method has strong robustness and is insensitive to the characteristics of the size, the direction and the like of an image. In the example, 6 statistics are used:
mean value:
variance:
skewness Skewness:
kurtosis:
energy:
entropy Encopy:
where h (i) is the gray level of each pixel, L is the number of pixel rows, σ is the standard deviation, and μ is the center distance. A CT image consists of pixels with gray levels from 0 to 255, while 0 represents black, which is usually the background of the image and is a useless part for feature extraction. The range of i is usually set to 1 to 255 in the calculation.
And the second method comprises the following steps: histogram of oriented gradient feature (HOG)
In feature extraction, not only statistical information of an image but also texture information is required because two images having the same statistical information may be completely different. To solve this problem, edge direction features in the HOG are used to describe the appearance and shape of the image. For the HOG algorithm, the image is first divided into connected regions called cells, and for each pixel in a cell, a histogram of directional gradients is used for compilation. To achieve higher accuracy, the normalized local histogram is contrasted by calculating the size of a larger region in the image, named the patch, and then this value can be used to normalize each cell. As shown in fig. 2, after the HOG features are extracted, a boundary containing edge direction information is drawn around the target, and finally a corresponding feature matrix is obtained.
The classifier model in the scheme is obtained by training different classifiers, and the specific process comprises the following steps:
s1021, collecting CT images of a plurality of known articles by using a security check instrument, and extracting features of each CT image to obtain the features of the CT images; in this step, the larger the number of CT images, the better the versatility of the final classifier model.
S1022, a data set is established, wherein the data set comprises a plurality of samples, and each sample comprises a characteristic corresponding to the known article and a label.
And S1023, normalizing the features in the data set, and then randomly dividing the data set into a training set and a test set.
And S1024, training different classifiers by using the training set to obtain classifier models, and testing the corresponding classifier models through the test set.
S1025, determining a classifier model from classifier models obtained by training different classifiers according to recall rate and accuracy as the preset classifier model; for example, selecting the classifier model with highest recall rate and highest accuracy; or preferentially select the classifier model with the highest accuracy.
In the embodiment of the present application, the different classifiers include an SVM classifier and a KNN classifier.
SVM classifier:
support vector machines SVMs were first proposed by Cottes and Vapnik in 1995. It exhibits many advantages in solving the low-sample, non-linear, high-dimensional pattern recognition problem and is applied to other machine learning problems such as function fitting.
The basic idea of SVM is to find a decision boundary with the best model generalization ability, so that it is as far as possible from all classified samples, that is, the sample points closest to the decision boundary in the samples are as far as possible from the decision boundary, and these points are the support vectors, and the dots shown in fig. 3 are the support vectors in this sample.
By means of these support vectors, two new straight lines can be defined again, there will be no data points in the area between these two straight lines, and the line in the middle of the area is the decision boundary we find, and also becomes the hyperplane. Assuming that all support vectors are at a distance d from the hyperplane, the distance margin between two support vector lines is 2 d. We further assume that the equation of a line of the hyperplane in n-dimensional space is:
w
Tx+b＝0
the distance from any point in space to the hyperplane is:
from the definition of the support vector, one can get:
and the objective function is:
namely, the following steps are required:
in the above formula, w represents a connection weight, b represents a straight line intercept, y represents an output, and i is a data point number.
It can be seen that this is a constrained optimization problem, and by using lagrange multipliers, the final optimal solution can be obtained.
In an embodiment, a library of SVM classifiers in the MATLAB statistical and machine learning toolkit is called, with the function name 'fitcsvm'. After reading all NIfTI format images, a corresponding labeled training set and a corresponding labeled testing set are introduced, and in order to obtain a better training effect, the data are normalized, and all characteristic values are converted into decimal numbers between (0, 1). And as the HOG features have tens of thousands of features, a Gaussian kernel function (rbf kernel) which is suitable for processing high-dimensional features in the SVM is selected for training the data. Finally, ten-fold cross validation is also performed on the training set to improve training accuracy.
KNN classifier:
the K-nearest neighbor algorithm selects the nearest K sample points around the sample point to be classified, and the class of the point is determined by the sample class with the largest number among the K samples.
In the embodiment, after the 'fitcknn' function in MATLAB is called and the training set and the test set are imported, the data also needs to be normalized. Meanwhile, the size of the neighbor K needs to be determined, and the default K-5 neighbor algorithm is selected. On the other hand, a proper distance description type is required to be selected, and after comparison, the Euclidean distance is finally selected as a classifier parameter. Finally, the data set is trained in the same fold-over cross validation manner.
Training results are as follows:
when using the grayscale features with the SVM classifier, the recall rate of the classifier model is about 43% and the accuracy is about 84%. For a particular identification target, the clay recall rate is highest and the rubber recall rate is lowest.
When the HOG features and the SVM classifier were used, the classifier model recall rate was about 42% and the accuracy was about 88%. For a particular identification target, clay and brine recall was highest and rubber recall was lowest.
When using the grayscale feature and KNN classifier, the recall rate is about 76% and the accuracy rate is about 31%. For a particular identification target, rubber recall is highest and brine recall is lowest.
Therefore, in the embodiment of the application, a classifier model obtained by training the SVM classifier can be selected.
According to another aspect of the present application, there is provided a device 1 for automatically identifying a target object by a security check machine based on machine learning, as shown in fig. 4, including:
the acquisition module 11 is configured to acquire a CT image of an object to be detected, which is acquired by the security inspection apparatus, and perform feature extraction on the CT image to obtain features of the CT image;
the recognition module 12 is configured to recognize features of the CT image by using a preset classifier model, and generate a recognition result corresponding to the object to be detected;
and the display module 13 is configured to send the identification result to the security check instrument, so that the security check instrument displays the identification result.
Further, the apparatus further comprises a classifier model determination module comprising:
the acquisition and extraction module is used for acquiring CT images of a plurality of known articles by using the security check instrument and extracting characteristics of each CT image to obtain the characteristics of the CT image;
a data set establishing module for establishing a data set, wherein the data set comprises a plurality of samples, and each sample comprises a characteristic corresponding to a known article and a label;
the data set dividing module is used for carrying out normalization processing on the features in the data set and then randomly dividing the data set into a training set and a testing set;
the training and testing module is used for training different classifiers by utilizing a training set to obtain classifier models and testing the corresponding classifier models through a testing set;
and the classifier determining module is used for determining one classifier model from classifier models obtained by training different classifiers as the preset classifier model according to the recall rate and the accuracy rate.
Referring to fig. 5, an embodiment of the present application further provides a terminal device 2, where the terminal device 2 may be a computer or a server; comprising a memory 22, a processor 21 and a computer program 23 stored in the memory 22 and operable on the processor, the processor 21, when executing the computer program 23, implements the above-mentioned method for automatically identifying a target object by a machine learning-based security check apparatus, for example, S101 to S103 shown in fig. 1.
The computer program 23 may also be divided into one or more modules/units, which are stored in the memory 22 and executed by the processor 21 to accomplish the present application. One or more modules/units may be a series of computer program instruction segments capable of performing specific functions, where the instruction segments are used to describe an execution process of the computer program 23 in the terminal device 2, for example, the computer program 23 may be divided into an obtaining module, an identifying module, and a displaying module, and functions of each module are described in the foregoing description, and are not repeated.
Implementations of the present application provide a computer-readable storage medium storing a computer program, which when executed by a processor implements the above-described method for automatically identifying a target object by a machine learning-based security check instrument, for example, S101 to S103 shown in fig. 1.
The integrated modules/units, if implemented in the form of software functional units and sold or used as separate products, may be stored in a computer readable storage medium. Based on such understanding, all or part of the flow in the method of the embodiments described above can be realized by a computer program, which can be stored in a computer readable storage medium and can realize the steps of the embodiments of the methods described above when the computer program is executed by a processor. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer readable medium may include: any entity or device capable of carrying computer program code, recording medium, U.S. disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-Only Memory (ROM), Random Access Memory (RAM), electrical carrier wave signals, telecommunications signals, software distribution media, and the like. It should be noted that the computer readable medium may contain other components which may be suitably increased or decreased as required by legislation and patent practice in jurisdictions, for example, in some jurisdictions, in accordance with legislation and patent practice, the computer readable medium does not include electrical carrier signals and telecommunications signals.
The above embodiments are only used to illustrate the technical solutions of the present application, and not to limit the same; 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 solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present application and are intended to be included within the scope of the present application.
Claims (10)
1. A method for automatically identifying a target object by a security check instrument based on machine learning is characterized by comprising the following steps:
acquiring a CT image of an article to be detected, which is acquired by a security inspection instrument, and performing feature extraction on the CT image to obtain features of the CT image;
identifying the characteristics of the CT image by using a preset classifier model to generate an identification result corresponding to the article to be detected;
and sending the identification result to a security check instrument so that the security check instrument displays the identification result.
2. The method for automatically identifying a target object by a machine learning-based security check machine according to claim 1, wherein before identifying the features of the CT image by using the preset classifier model, the method further comprises:
the features of the CT image are normalized to convert all feature values to values between 0 and 1.
3. The method for automatically identifying a target object by a machine learning-based security check machine according to claim 1, wherein the training process of the preset classifier model comprises:
acquiring CT images of a plurality of known articles by using a security check instrument, and extracting characteristics of each CT image to obtain the characteristics of the CT images;
establishing a data set, wherein the data set comprises a plurality of samples, and each sample comprises a characteristic corresponding to a known article and a label;
normalizing the characteristics in the data set, and then randomly dividing the data set into a training set and a testing set;
training different classifiers by using a training set to obtain classifier models, and testing the corresponding classifier models by using a test set;
and determining a classifier model from classifier models obtained by training different classifiers as the preset classifier model according to the recall rate and the accuracy.
4. The method for automatically identifying a target object by a machine learning-based security check instrument as claimed in claim 1, wherein the feature extraction is performed on the CT image, wherein the extracted features are a gray scale feature and a histogram of oriented gradients feature of the CT image.
5. The method for machine learning-based security check instrument automatic target object identification of claim 1, wherein the different classifiers include SVM classifier and KNN classifier.
6. The method for automatically identifying a target object by a machine learning-based security check instrument as claimed in claim 5, wherein when the classifier is an SVM classifier, a Gaussian kernel function is adopted for training in the training process, and cross validation is performed on a training set by ten folds.
7. The method for automatically identifying a target object by a machine learning-based security check machine according to claim 5, wherein when the classifier is a KNN classifier, the value of K is selected to be 5; and (5) taking the Euclidean distance as a parameter of the KNN classifier, and performing ten-fold cross validation on the training set.
8. A device for automatically identifying a target object by a security check instrument based on machine learning is characterized by comprising:
the acquisition module is used for acquiring a CT image of the to-be-detected article acquired by the security check instrument and extracting the characteristics of the CT image to obtain the characteristics of the CT image;
the recognition module is used for recognizing the characteristics of the CT image by utilizing a preset classifier model and generating a recognition result corresponding to the article to be detected;
and the display module is used for sending the identification result to a security check instrument so that the security check instrument displays the identification result.
9. A terminal device comprising a memory, a processor and a computer program stored in said memory and executable on said processor, characterized in that the processor implements the steps of the method for automatically identifying a target object by a machine learning based security check instrument of the aforementioned first aspect when executing the computer program.
10. A computer-readable storage medium, in which a computer program is stored, wherein the computer program, when being executed by a processor, implements the steps of the method for automatically identifying a target object by a machine learning-based security check instrument according to the first aspect.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911017535.9A CN110781812A (en) | 2019-10-24 | 2019-10-24 | Method for automatically identifying target object by security check instrument based on machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911017535.9A CN110781812A (en) | 2019-10-24 | 2019-10-24 | Method for automatically identifying target object by security check instrument based on machine learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110781812A true CN110781812A (en) | 2020-02-11 |
Family
ID=69387358
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911017535.9A Pending CN110781812A (en) | 2019-10-24 | 2019-10-24 | Method for automatically identifying target object by security check instrument based on machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110781812A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113362273A (en) * | 2020-02-19 | 2021-09-07 | 大众汽车股份公司 | Method, system and computer program for processing computed tomography image data of a component and training a machine learning model |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106250936A (en) * | 2016-08-16 | 2016-12-21 | 广州麦仑信息科技有限公司 | Multiple features multithreading safety check contraband automatic identifying method based on machine learning |
| CN107871122A (en) * | 2017-11-14 | 2018-04-03 | 深圳码隆科技有限公司 | Safety check detection method, device, system and electronic equipment |
| CN108198227A (en) * | 2018-03-16 | 2018-06-22 | 济南飞象信息科技有限公司 | Contraband intelligent identification Method based on X-ray screening machine image |
| CN109344905A (en) * | 2018-10-22 | 2019-02-15 | 王子蕴 | A kind of transmission facility automatic fault recognition methods based on integrated study |
| CN109347719A (en) * | 2018-09-11 | 2019-02-15 | 内蒙古工业大学 | A kind of image junk mail filtering method based on machine learning |
| CN109902643A (en) * | 2019-03-07 | 2019-06-18 | 浙江啄云智能科技有限公司 | Intelligent safety inspection method, device, system and its electronic equipment based on deep learning |
| CN109975335A (en) * | 2019-03-07 | 2019-07-05 | 北京航星机器制造有限公司 | A kind of CT detection method and device |
| CN110363812A (en) * | 2019-07-10 | 2019-10-22 | 国网四川省电力公司电力科学研究院 | A kind of image-recognizing method |
-
2019
- 2019-10-24 CN CN201911017535.9A patent/CN110781812A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106250936A (en) * | 2016-08-16 | 2016-12-21 | 广州麦仑信息科技有限公司 | Multiple features multithreading safety check contraband automatic identifying method based on machine learning |
| CN107871122A (en) * | 2017-11-14 | 2018-04-03 | 深圳码隆科技有限公司 | Safety check detection method, device, system and electronic equipment |
| WO2019096181A1 (en) * | 2017-11-14 | 2019-05-23 | 深圳码隆科技有限公司 | Detection method, apparatus and system for security inspection, and electronic device |
| CN108198227A (en) * | 2018-03-16 | 2018-06-22 | 济南飞象信息科技有限公司 | Contraband intelligent identification Method based on X-ray screening machine image |
| CN109347719A (en) * | 2018-09-11 | 2019-02-15 | 内蒙古工业大学 | A kind of image junk mail filtering method based on machine learning |
| CN109344905A (en) * | 2018-10-22 | 2019-02-15 | 王子蕴 | A kind of transmission facility automatic fault recognition methods based on integrated study |
| CN109902643A (en) * | 2019-03-07 | 2019-06-18 | 浙江啄云智能科技有限公司 | Intelligent safety inspection method, device, system and its electronic equipment based on deep learning |
| CN109975335A (en) * | 2019-03-07 | 2019-07-05 | 北京航星机器制造有限公司 | A kind of CT detection method and device |
| CN110363812A (en) * | 2019-07-10 | 2019-10-22 | 国网四川省电力公司电力科学研究院 | A kind of image-recognizing method |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113362273A (en) * | 2020-02-19 | 2021-09-07 | 大众汽车股份公司 | Method, system and computer program for processing computed tomography image data of a component and training a machine learning model |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Seeland et al. | Plant species classification using flower images—A comparative study of local feature representations | |
| CN107239786B (en) | Character recognition method and device | |
| US9418283B1 (en) | Image processing using multiple aspect ratios | |
| JP5570629B2 (en) | Classifier learning method and apparatus, and processing apparatus | |
| EP2743888B1 (en) | Feature extraction device, feature extraction program, and image processing device | |
| US8699789B2 (en) | Document classification using multiple views | |
| JP6897749B2 (en) | Learning methods, learning systems, and learning programs | |
| CN111400528B (en) | Image compression method, device, server and storage medium | |
| CN113269257A (en) | Image classification method and device, terminal equipment and storage medium | |
| CN115346227B (en) | Method for vectorizing electronic file based on layout file | |
| CN111860536A (en) | Image recognition method, device and storage medium | |
| KR101232683B1 (en) | Method for recognizing denomination of banknotes using Bayesian approach | |
| US9058517B1 (en) | Pattern recognition system and method using Gabor functions | |
| CN115171125A (en) | Data anomaly detection method | |
| CN117315377B (en) | Image processing method and device based on machine vision and electronic equipment | |
| CN113269752A (en) | Image detection method, device terminal equipment and storage medium | |
| CN110781812A (en) | Method for automatically identifying target object by security check instrument based on machine learning | |
| CN111476226B (en) | Text positioning method and device and model training method | |
| CN111401485A (en) | Practical texture classification method | |
| KR20220001918A (en) | Apparatus and method for phase segmentation of steel microstructure | |
| US20160358039A1 (en) | Apparatus and method for detecting object | |
| Mohammed et al. | Proposed approach for automatic underwater object classification | |
| Jubair et al. | A simplified method for handwritten character recognition from document image | |
| CN113177602B (en) | Image classification method, device, electronic equipment and storage medium | |
| CN113537158B (en) | Image target detection method, device, equipment and storage medium |
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 |