CN107871122A - Safety check detection method, device, system and electronic equipment - Google Patents

Safety check detection method, device, system and electronic equipment Download PDF

Info

Publication number
CN107871122A
CN107871122A CN201711126618.2A CN201711126618A CN107871122A CN 107871122 A CN107871122 A CN 107871122A CN 201711126618 A CN201711126618 A CN 201711126618A CN 107871122 A CN107871122 A CN 107871122A
Authority
CN
China
Prior art keywords
safety check
ray image
deep learning
ray
recognition result
Prior art date
Application number
CN201711126618.2A
Other languages
Chinese (zh)
Inventor
黄鼎隆
马修·罗伯特·斯科特
董登科
王重
Original Assignee
深圳码隆科技有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 深圳码隆科技有限公司 filed Critical 深圳码隆科技有限公司
Priority to CN201711126618.2A priority Critical patent/CN107871122A/en
Publication of CN107871122A publication Critical patent/CN107871122A/en

Links

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01VGEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS
    • G01V5/00Prospecting or detecting by the use of nuclear radiation, e.g. of natural or induced radioactivity
    • G01V5/0008Detecting hidden objects, e.g. weapons, explosives
    • G01V5/0016Active interrogation, i.e. using an external radiation source, e.g. using pulsed, continuous or cosmic rays
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K9/00Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
    • G06K9/00624Recognising scenes, i.e. recognition of a whole field of perception; recognising scene-specific objects
    • G06K9/00664Recognising scenes such as could be captured by a camera operated by a pedestrian or robot, including objects at substantially different ranges from the camera
    • G06K9/00671Recognising scenes such as could be captured by a camera operated by a pedestrian or robot, including objects at substantially different ranges from the camera for providing information about objects in the scene to a user, e.g. as in augmented reality applications
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K9/00Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
    • G06K9/62Methods or arrangements for recognition using electronic means
    • G06K9/6217Design or setup of recognition systems and techniques; Extraction of features in feature space; Clustering techniques; Blind source separation
    • G06K9/6256Obtaining sets of training patterns; Bootstrap methods, e.g. bagging, boosting

Abstract

The invention provides a kind of safety check detection method, device, system and electronic equipment, wherein, this method includes obtaining the x-ray image of the X-ray machine collection in the screening machine that safety check terminal receives, and the x-ray image is pre-processed, obtains pretreated x-ray image;According to the article characteristics of corresponding thing to be detected in the default pretreated x-ray image of deep learning model extraction, the default deep learning model includes the deep learning model based on convolutional neural networks;Article characteristics are identified using the grader based on default deep learning model training, the recognition result of the corresponding thing to be detected of generation;The recognition result of thing to be detected is sent to safety check terminal, so that safety check terminal shows the recognition result.Technical scheme provided in an embodiment of the present invention, the automatic identification detection to contraband is realized, and while recognition efficiency is improved, the accuracy to contraband identification has been effectively ensured, has prevented the generation of potential safety hazard.

Description

Safety check detection method, device, system and electronic equipment

Technical field

The present invention relates to technical field of security inspection equipment, more particularly, to a kind of safety check detection method, device, system and electronics Equipment.

Background technology

With the enhancing that public security is realized, various rays safety detection apparatus are widely used in airport, port, harbour, subway, method Institute, the important public place such as venue of important sports events.And most common of which is exactly (X ray) screening machine, screening machine exists at present Passenger flow and logistics field have obtained more and more extensive application.General screening machine all includes motor, by motor-driven conveyer belt System, the fuselage being crossed in the middle part of conveyer belt system, certainly also supporting X-ray machine.The fuselage and conveyer belt of usual screening machine Between form security check passage, at least side of security check passage is provided with X-ray machine.One end of conveyer belt system is region to be checked, to be checked After survey thing is placed on region to be checked, the conveyer belt system that can be motor driven is transferred to the other end, necessarily passes safety check halfway and leads to Road, scanning generation X-ray imaging is irradiated by X-ray machine, so as to identify whether thing to be detected is contraband.

But existing detection method is due to easily by the outside environmental elements such as detection penetrability, detection angles or outer Boundary's interference effect, greatly reduce identification accuracy;And, it is necessary to which security staff is to the figure of display after terminal X light image Piece is investigated.So security staff's long-time monitor screen easily causes visual fatigue, causes the feelings such as flase drop, false retrieval, missing inspection Condition occurs.

Therefore, existing safety check detection method, it is difficult to ensure the accuracy of identification, and recognition efficiency is low, easily causes peace Full hidden danger.

The content of the invention

In view of this, it is an object of the invention to provide a kind of safety check detection method, device, system and electronic equipment, with While recognition efficiency is improved, the accuracy to contraband identification is effectively ensured, prevents potential safety hazard.

In a first aspect, the embodiments of the invention provide a kind of safety check detection method, including:

The x-ray image of the X-ray machine collection in the screening machine that safety check terminal receives is obtained, the x-ray image is carried out pre- Processing, obtains pretreated x-ray image;

According to the article of corresponding thing to be detected in pretreated x-ray image described in default deep learning model extraction Feature, the default deep learning model include the deep learning model based on convolutional neural networks;

The article characteristics are identified using the grader based on the default deep learning model training, generated The recognition result of the corresponding thing to be detected;

The recognition result of the thing to be detected is sent to the safety check terminal, so that the safety check terminal shows the knowledge Other result.

With reference in a first aspect, the embodiments of the invention provide the possible embodiment of the first of first aspect, wherein, institute State and the x-ray image is pre-processed, including:

Smoothing denoising is carried out to the x-ray image of collection using neighborhood averaging, obtains the X-ray figure after smoothing denoising Picture;

The marginal information of the image after the smoothing denoising is strengthened using histogram equalization method, after obtaining pretreatment X-ray image.

With reference in a first aspect, the embodiments of the invention provide the possible embodiment of second of first aspect, wherein, institute It is that the article sample data for exceeding certain threshold value by quantity trains to obtain to state the deep learning model based on convolutional neural networks , the article sample data includes picture corresponding to the contraband of different shape.

With reference in a first aspect, the embodiments of the invention provide the possible embodiment of the third of first aspect, wherein, institute Stating the training process of grader includes:

Utilize the depth characteristic of the deep learning model extraction article sample data based on convolutional neural networks;

Based on machine learning algorithm, grader is trained to the depth characteristic;

Wherein described article sample data include the X-ray picture for the different recognition results specified.

Second aspect, the embodiment of the present invention also provide a kind of safety check detection means, including:

Pretreatment module, the x-ray image of the X-ray machine collection in the screening machine received for obtaining safety check terminal, to institute State x-ray image to be pre-processed, obtain pretreated x-ray image;

Characteristic extracting module, for right in the pretreated x-ray image according to default deep learning model extraction The article characteristics for the thing to be detected answered, the default deep learning model include the deep learning mould based on convolutional neural networks Type;

As a result identification module, for utilizing the grader based on the default deep learning model training to the article Feature is identified, the recognition result of the corresponding thing to be detected of generation;

Result display module, for the recognition result of the thing to be detected to be sent to the safety check terminal, so that described Safety check terminal shows the recognition result.

The third aspect, the embodiment of the present invention also provide a kind of safety check detecting system, including screening machine, safety check terminal and safety check Identification equipment, X-ray machine is provided with the safety check case of the screening machine, and the safety check identification equipment is included as described in second aspect Safety check detection means;The X-ray machine, the safety check identification equipment are connected with the safety check terminal respectively;

The X-ray machine, the x-ray image of the thing to be detected for gathering the security check passage by the screening machine, by the X Light image is sent to the safety check terminal;

The safety check terminal, for when listening to the x-ray image of reception, the x-ray image to be sent to safety check Identification equipment;It is additionally operable to receive the recognition result for the thing to be detected that the safety check identification equipment is sent, the identification is tied Fruit is shown by display screen.

With reference to the third aspect, the embodiments of the invention provide the possible embodiment of the first of the third aspect, wherein, institute Stating recognition result includes contraband and non-contraband two types;The system also includes warning device, the warning device with The safety check terminal connection;

The safety check terminal, it is additionally operable to when the recognition result received is contraband, sends alarm signal to institute Warning device is stated, so that the warning device carries out alarm.

With reference to the third aspect, the embodiments of the invention provide the possible embodiment of second of the third aspect, wherein, institute The bottom for stating the security check passage of screening machine is provided with pressure sensor, and the pressure sensor is connected with the safety check terminal;

The pressure sensor, for gathering the pressure information born on the security check passage, the pressure information is sent out Deliver to the safety check terminal;

The safety check terminal, it is additionally operable to be turned on or off the X-ray machine according to the pressure information.

With reference to the third aspect and its any possible embodiment, the embodiments of the invention provide the 3rd of the third aspect The possible embodiment of kind, wherein, the safety check identification equipment includes Nvidia Jetson TX2 chips.

Fourth aspect, the embodiment of the present invention also provide a kind of electronic equipment, including memory, processor, the memory On be stored with the computer program that can be run on the processor, described in the computing device during computer program realize on State first aspect and its method described in any possible embodiment.

The embodiment of the present invention brings following beneficial effect:

The embodiments of the invention provide a kind of safety check detection method, device, system and electronic equipment, wherein this method includes The x-ray image of the X-ray machine collection in the screening machine that safety check terminal receives is obtained, the x-ray image is pre-processed, obtained pre- X-ray image after processing;According to corresponding to be detected in pretreated x-ray image described in default deep learning model extraction The article characteristics of thing, the default deep learning model include the deep learning model based on convolutional neural networks;Using based on Article characteristics are identified the grader of default deep learning model training, the recognition result of the corresponding thing to be detected of generation; The recognition result of thing to be detected is sent to safety check terminal, so that safety check terminal shows the recognition result.In the embodiment of the present invention In the technical scheme of offer, reduce the influence of outside environmental elements first with pretreatment, then by based on convolutional Neural net The deep learning model extraction article characteristics of network, and treat detectable substance using the grader of the deep learning model training and known Not, the automatic identification detection to contraband has been achieved in that, and while recognition efficiency is improved, has been effectively ensured to contraband The accuracy of identification, prevent the generation of potential safety hazard.

Other features and advantages of the present invention will illustrate in the following description, also, partly become from specification Obtain it is clear that or being understood by implementing the present invention.The purpose of the present invention and other advantages are in specification, claims And specifically noted structure is realized and obtained in accompanying drawing.

To enable the above objects, features and advantages of the present invention to become apparent, preferred embodiment cited below particularly, and coordinate Appended accompanying drawing, is described in detail below.

Brief description of the drawings

, below will be to specific in order to illustrate more clearly of the specific embodiment of the invention or technical scheme of the prior art The required accompanying drawing used is briefly described in embodiment or description of the prior art, it should be apparent that, in describing below Accompanying drawing is some embodiments of the present invention, for those of ordinary skill in the art, before creative work is not paid Put, other accompanying drawings can also be obtained according to these accompanying drawings.

Fig. 1 is the schematic flow sheet of safety check detection method provided in an embodiment of the present invention;

Fig. 2 is the structural representation of safety check detection means provided in an embodiment of the present invention;

Fig. 3 is the communication link map interlinking of safety check detecting system provided in an embodiment of the present invention;

Fig. 4 is the structural representation of electronic equipment provided in an embodiment of the present invention.

Embodiment

To make the purpose, technical scheme and advantage of the embodiment of the present invention clearer, below in conjunction with accompanying drawing to the present invention Technical scheme be clearly and completely described, it is clear that described embodiment is part of the embodiment of the present invention, rather than Whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art are not making creative work premise Lower obtained every other embodiment, belongs to the scope of protection of the invention.

Current existing safety check detection method, it is difficult to ensure the accuracy of identification, and recognition efficiency is low, easily causes safety Hidden danger, based on this, a kind of safety check detection method, device, system and electronic equipment provided in an embodiment of the present invention, it can utilize pre- Processing reduces the influence of outside environmental elements, then special by the deep learning model extraction article based on convolutional neural networks Sign, and treat detectable substance using the grader of the deep learning model training and be identified, be achieved in that to contraband from Dynamic recognition detection, and while recognition efficiency is improved, the accuracy to contraband identification has been effectively ensured, prevent safety hidden The generation of trouble.

For ease of understanding the present embodiment, a kind of safety check detection method disclosed in the embodiment of the present invention is entered first Row is discussed in detail.

Embodiment one:

Safety check detection method provided in an embodiment of the present invention can be, but not limited to be applied to airport, port, harbour, subway, Law court, important sports events the safety check scene of important public place such as venue in.

Fig. 1 shows the schematic flow sheet of safety check detection method provided in an embodiment of the present invention.As shown in figure 1, the safety check Detection method includes:

Step S101, the x-ray image of the X-ray machine collection in the screening machine that safety check terminal receives is obtained, to the x-ray image Pre-processed, obtain pretreated x-ray image.

The safety check terminal can be, but not limited to as computer, console.Specifically, when product to be detected pass through the safety check of screening machine During passage, the X-ray machine being arranged in the safety check case of screening machine can be scanned to product to be detected and generate x-ray image, by the X-ray Image is sent to above-mentioned safety check terminal.

The x-ray image of safety check terminal transmission is being got, during detecting x-ray image, is requiring that the x-ray image has first There is good performance.But because external interference and X-ray machine oneself factor can cause the reduction of x-ray image quality, based on this The pretreatment operation that inventive embodiments use mainly includes image enhaucament and denoising, above-mentioned to carry out pretreatment bag to the x-ray image Include:

Smoothing denoising is carried out to the x-ray image of collection using neighborhood averaging, obtains the x-ray image after smoothing denoising;

The marginal information of the image after smoothing denoising is strengthened using histogram equalization method, obtains pretreated X Light image.

By above-mentioned preprocess method, the useful information such as marginal information in image can be strengthened, to a certain extent Weaken interference (medium scatters, high-speed motion, noise jamming), improve the performance of image so that the feature of image is fully shown Come, be more beneficial for the feature extraction and expression in later stage.

Step S102, according to corresponding to be checked in the above-mentioned pretreated x-ray image of default deep learning model extraction The article characteristics of thing are surveyed, the default deep learning model includes the deep learning model based on convolutional neural networks.

The above-mentioned deep learning model based on convolutional neural networks is the article sample number for exceeding certain threshold value by quantity Obtained according to training, the article sample data includes X-ray picture corresponding to the contraband of different shape.Wherein, contraband is not Include the state that such as gun split into the state of each parts, cutter folds with form.In a preferred embodiment, The above-mentioned deep learning model based on convolutional neural networks can be realized by Caffe deep learnings framework.

Specifically, the quantity of above-mentioned X-ray picture is The more the better, and data are more, train generation based on convolutional neural networks Deep learning model versatility it is better, such as above-mentioned X-ray picture include multiple angles, the various contrabands of a variety of penetration levels Picture, be so advantageous to subsequently treat accurately identifying for detectable substance, overcome the influence of outside environmental elements, improving should be based on volume The recognition capability of the deep learning model of product neutral net.

Step S102 is specifically included:Using pretreated x-ray image as input picture in default deep learning mould Carry out features training successively in the multiple basic units included in type, after the completion of training, extract it is multiple it is integrated in full articulamentum or Person other specify the characteristic vector of basic units' output as corresponding to article characteristics to be detected in pretreated x-ray image.

Further, in order to simple and handle characteristics of image rapidly, in step s 102 according to default deep learning mould Type is extracted in pretreated x-ray image before the article characteristics of corresponding thing to be detected, in addition to:

Pretreated x-ray image is divided into according to the gamma characteristic of pretreated x-ray image by background and the class of template two, The variance made between two classes is obtained into maximum parameter as optimal threshold;

Binary image is obtained using the Optimal-threshold segmentation, using the binary image as pretreated x-ray image.

In various threshold optimization dividing methods, OTSU algorithms propose that maximizing split plot design based on inter-class variance is acknowledged as It is Optimal-threshold segmentation algorithm, it divides the image into background and the class of target two according to the gamma characteristic of image, then calculates and allows two Variance between class obtains maximum parameter as optimal threshold, the binary picture for recycling Optimal-threshold segmentation to be worked well Picture.

Thus, by above-mentioned binary conversion treatment, the gray value of the pixel on x-ray image is arranged to 0 or 255, X-ray Data volume is greatly reduced in image, and image processing speed can substantially reduce.

Step S103, above-mentioned article characteristics are carried out using the grader based on above-mentioned default deep learning model training Identification, the recognition result of the corresponding thing to be detected of generation.

Input i.e. using the feature extracted in step 102 as the grader based on default deep learning model training, After being identified by the grader, final recognition result is obtained.Specifically, recognition result is contraband or non-contraband, can With but be not limited by correct or error identification, and apply different picture identifications, be specifically identified method and be not construed as limiting here.

In an optional embodiment, the training process for the grader applied in step 103 includes:

Utilize the depth characteristic of the deep learning model extraction article sample data based on convolutional neural networks;

Based on machine learning algorithm, grader is trained to above-mentioned depth characteristic;

Wherein above-mentioned article sample data include the X-ray picture for the different recognition results specified.Above-mentioned machine learning is calculated Method can be nearest neighbor algorithm, EM algorithm and algorithm of support vector machine etc., and specific algorithm can select as the case may be, Here it is not construed as limiting.

In an optional embodiment, above-mentioned article sample data include triple data;The wherein triple data Including:Source data and source data belong to same category of forward data and adhere to different classes of reverse number separately with the source data According to.

Wherein, source data is the recognition result identical sample data got at random from article sample data.

Forward data is the sample data consistent with the recognition result of source data obtained at random from article sample data; The matching degree of the source data is higher than the matching degree of forward data.

Reverse data is the sample number inconsistent with the recognition result of source data obtained at random from article sample data According to.

In a specific embodiment, triple data are respectively:X-ray image is of good performance in article sample data First picture (source data), the second picture (forward data) of the x-ray image poor-performing shot in article sample data, and The 3rd picture as reverse data different from the first picture and second picture recognition result.First picture and second picture Recognition result is contraband, and the recognition result of second picture is non-contraband.Second picture is because image property is poor, such as clear There is gap in clear degree, resolution ratio etc., its matching degree is less than the first picture with the first picture.3rd picture is then in training The reverse data of reverse contrast is carried out, once by positive and negative contrast, further enhancing the recognition capability of grader.

Step S104, the recognition result of above-mentioned thing to be detected is sent to safety check terminal, so that safety check terminal shows the knowledge Other result.

Specifically, the safety check terminal is carried out after the recognition result of thing to be detected is received in the display interface of display screen Render, to show the recognition result.

In technical scheme provided in an embodiment of the present invention, reduce the influence of outside environmental elements first with pretreatment, Then by the deep learning model extraction article characteristics based on convolutional neural networks, and the deep learning model training is utilized Grader is treated detectable substance and is identified, and has been achieved in that the automatic identification detection to contraband, and improving recognition efficiency Meanwhile the accuracy to contraband identification has been effectively ensured, prevent the generation of potential safety hazard.

Embodiment two:

Fig. 2 shows the structural representation of safety check detection means provided in an embodiment of the present invention.As shown in Fig. 2 the safety check Detection means includes:

Pretreatment module 11, the x-ray image of the X-ray machine collection in the screening machine received for obtaining safety check terminal are right The x-ray image is pre-processed, and obtains pretreated x-ray image;

Characteristic extracting module 12, for according in the above-mentioned pretreated x-ray image of default deep learning model extraction The article characteristics of corresponding thing to be detected, the default deep learning model include the deep learning mould based on convolutional neural networks Type;

As a result identification module 13, for utilizing the grader based on above-mentioned default deep learning model training to above-mentioned thing Product feature is identified, the recognition result of the corresponding thing to be detected of generation;

Result display module 14, for the recognition result of above-mentioned thing to be detected to be sent to safety check terminal, so that safety check is whole End shows the recognition result.

The above-mentioned deep learning model based on convolutional neural networks is the article sample number for exceeding certain threshold value by quantity Obtained according to training, the article sample data includes X-ray picture corresponding to the contraband of different shape.Wherein, contraband is not Include the state that such as gun split into the state of each parts, cutter folds with form.In a preferred embodiment, The above-mentioned deep learning model based on convolutional neural networks can be realized by Caffe deep learnings framework.

In technical scheme provided in an embodiment of the present invention, reduce the influence of outside environmental elements first with pretreatment, Then by the deep learning model extraction article characteristics based on convolutional neural networks, and the deep learning model training is utilized Grader is treated detectable substance and is identified, and has been achieved in that the automatic identification detection to contraband, and improving recognition efficiency Meanwhile the accuracy to contraband identification has been effectively ensured, prevent the generation of potential safety hazard.

Embodiment three:

Fig. 3 shows the communication link map interlinking of safety check detecting system provided in an embodiment of the present invention.As shown in figure 3, the safety check Detecting system includes:Including screening machine 400, safety check terminal 500 and safety check identification equipment 600, set in the safety check case of the screening machine X-ray machine 700 is equipped with, safety check identification equipment is included such as the safety check detection means in embodiment two;X-ray machine, safety check identification equipment point It is not connected with safety check terminal.

X-ray machine, the x-ray image of the thing to be detected for gathering the security check passage by screening machine, the x-ray image is sent To the safety check terminal.

Safety check terminal, for when listening to the above-mentioned x-ray image of reception, x-ray image to be sent to safety check identification equipment; It is additionally operable to receive the recognition result for the thing to be detected that safety check identification equipment is sent, the recognition result is shown by display screen.

In technical scheme provided in an embodiment of the present invention, reduce the influence of outside environmental elements first with pretreatment, Then by the deep learning model extraction article characteristics based on convolutional neural networks, and the deep learning model training is utilized Grader is treated detectable substance and is identified, and has been achieved in that the automatic identification detection to contraband, and improving recognition efficiency Meanwhile the accuracy to contraband identification has been effectively ensured, prevent the generation of potential safety hazard.

In an optional embodiment, above-mentioned recognition result includes contraband and non-contraband two types;Above-mentioned peace Inspection detecting system also includes warning device 800, and the warning device is connected with safety check terminal.

Specifically, the safety check terminal is additionally operable to when the above-mentioned recognition result received is contraband, sends alarm signal To warning device, so that warning device carries out alarm.Wherein the type of alarm of warning device includes light warning, voice reporting The alarm of alert or picture and text showing.

In another optional embodiment, the bottom of the security check passage of screening machine is provided with pressure sensor 900, the pressure Force snesor is connected with safety check terminal.

Pressure sensor is used to gather the pressure information born on security check passage, and pressure information is sent to safety check terminal. Safety check terminal is additionally operable to be turned on or off X-ray machine according to the pressure information.

Specifically, the pressure value of pressure sensor is read when security check passage zero load, using the pressure value as pressure Threshold value, when the pressure information that pressure sensor that safety check terminal receives is sent exceedes the pressure threshold, illustrate to have to be detected Thing will then open X-ray machine by security check passage, so that the X-ray machine gathers the x-ray image of thing to be detected.When safety check terminal connects When the pressure information that the pressure sensor received is sent returns to the pressure threshold, illustrate thing to be detected from security check passage It is removed, closes the X-ray machine.So, automatically opening up and closing for X-ray machine is realized, serves and saves the energy and extension machine The effect of service life.

In one embodiment, above-mentioned safety check identification equipment includes Nvidia Jetson TX2 chips, and the chip is in low work( Powerful operational capability is maintained while consumption, it is possible to achieve the prohibited items in Millisecond identifies X-ray picture.Whole core Only credit card-sized, the Real time identification of contraband can be both realized on the basis of not transforming existing X-ray machine, again can be with X-ray machine combines offer identification service.

Example IV:

Referring to Fig. 4, the embodiment of the present invention also provides a kind of electronic equipment 100, including:Processor 40, memory 41, bus 42 and communication interface 43, the processor 40, communication interface 43 and memory 41 connected by bus 42;Processor 40 is used to hold The executable module stored in line storage 41, such as computer program.

Wherein, memory 41 may include high-speed random access memory (RAM, Random Access Memory), Non-labile memory (non-volatile memory), for example, at least a magnetic disk storage may also be included.By extremely A few communication interface 43 (can be wired or wireless) is realized logical between the system network element and at least one other network element Letter connection, can use internet, wide area network, LAN, Metropolitan Area Network (MAN) etc..

Bus 42 can be isa bus, pci bus or eisa bus etc..The bus can be divided into address bus, data Bus, controlling bus etc..Only represented for ease of representing, in Fig. 4 with a four-headed arrow, it is not intended that an only bus or A type of bus.

Wherein, memory 41 is used for storage program, and the processor 40 performs the journey after execute instruction is received Sequence, the method performed by device that the stream process that foregoing any embodiment of the embodiment of the present invention discloses defines can apply to handle In device 40, or realized by processor 40.

Processor 40 is probably a kind of IC chip, has the disposal ability of signal.In implementation process, above-mentioned side Each step of method can be completed by the integrated logic circuit of the hardware in processor 40 or the instruction of software form.Above-mentioned Processor 40 can be general processor, including central processing unit (Central Processing Unit, abbreviation CPU), network Processor (Network Processor, abbreviation NP) etc.;It can also be digital signal processor (Digital Signal Processing, abbreviation DSP), application specific integrated circuit (Application Specific Integrated Circuit, referred to as ASIC), ready-made programmable gate array (Field-Programmable Gate Array, abbreviation FPGA) or other are programmable Logical device, discrete gate or transistor logic, discrete hardware components.It can realize or perform in the embodiment of the present invention Disclosed each method, step and logic diagram.General processor can be microprocessor or the processor can also be appointed What conventional processor etc..The step of method with reference to disclosed in the embodiment of the present invention, can be embodied directly in hardware decoding processing Device performs completion, or performs completion with the hardware in decoding processor and software module combination.Software module can be located at Machine memory, flash memory, read-only storage, programmable read only memory or electrically erasable programmable memory, register etc. are originally In the ripe storage medium in field.The storage medium is located at memory 41, and processor 40 reads the information in memory 41, with reference to Its hardware completes the step of above method.

Safety check detection means, system and electronic equipment provided in an embodiment of the present invention, the safety check provided with above-described embodiment Detection method has identical technical characteristic, so can also solve identical technical problem, reaches identical technique effect.

The computer program product for the progress safety check detection method that the embodiment of the present invention is provided, including store processor The computer-readable recording medium of executable non-volatile program code, the instruction that described program code includes can be used for performing Method described in previous methods embodiment, specific implementation can be found in embodiment of the method, will not be repeated here.

It is apparent to those skilled in the art that for convenience and simplicity of description, the device of foregoing description, The specific work process of system and electronic equipment, the corresponding process in preceding method embodiment is may be referred to, will not be repeated here.

Flow chart and block diagram in accompanying drawing show multiple embodiment method and computer program products according to the present invention Architectural framework in the cards, function and operation.At this point, each square frame in flow chart or block diagram can represent one A part for module, program segment or code, a part for the module, program segment or code include one or more and are used to realize The executable instruction of defined logic function.It should also be noted that at some as the work(in the realization replaced, marked in square frame Energy can also be with different from the order marked in accompanying drawing generation.For example, two continuous square frames can essentially be substantially parallel Ground is performed, and they can also be performed in the opposite order sometimes, and this is depending on involved function.It is also noted that block diagram And/or the combination of each square frame and block diagram in flow chart and/or the square frame in flow chart, work(as defined in performing can be used Can or the special hardware based system of action realize, or the combination of specialized hardware and computer instruction can be used come reality It is existing.In addition, term " first ", " second ", " the 3rd " are only used for describing purpose, and it is not intended that instruction or implying relatively important Property.

In several embodiments provided herein, it should be understood that disclosed systems, devices and methods, can be with Realize by another way.Device embodiment described above is only schematical, for example, the division of the unit, Only a kind of division of logic function, can there is other dividing mode when actually realizing, in another example, multiple units or component can To combine or be desirably integrated into another system, or some features can be ignored, or not perform.It is another, it is shown or beg for The mutual coupling of opinion or direct-coupling or communication connection can be by some communication interfaces, device or unit it is indirect Coupling or communication connection, can be electrical, mechanical or other forms.

The unit illustrated as separating component can be or may not be physically separate, show as unit The part shown can be or may not be physical location, you can with positioned at a place, or can also be distributed to multiple On NE.Some or all of unit therein can be selected to realize the mesh of this embodiment scheme according to the actual needs 's.

In addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, can also That unit is individually physically present, can also two or more units it is integrated in a unit.

If the function is realized in the form of SFU software functional unit and is used as independent production marketing or in use, can be with It is stored in the executable non-volatile computer read/write memory medium of a processor.Based on such understanding, the present invention The part that is substantially contributed in other words to prior art of technical scheme or the part of the technical scheme can be with software The form of product is embodied, and the computer software product is stored in a storage medium, including some instructions are causing One computer equipment (can be personal computer, server, or network equipment etc.) performs each embodiment institute of the present invention State all or part of step of method.And foregoing storage medium includes:USB flash disk, mobile hard disk, read-only storage (ROM, Read- Only Memory), random access memory (RAM, Random Access Memory), magnetic disc or CD etc. are various can be with The medium of store program codes.

Finally it should be noted that:Embodiment described above, it is only the embodiment of the present invention, to illustrate the present invention Technical scheme, rather than its limitations, protection scope of the present invention is not limited thereto, although with reference to the foregoing embodiments to this hair It is bright to be described in detail, it will be understood by those within the art that:Any one skilled in the art The invention discloses technical scope in, it can still modify to the technical scheme described in previous embodiment or can be light Change is readily conceivable that, or equivalent substitution is carried out to which part technical characteristic;And these modifications, change or replacement, do not make The essence of appropriate technical solution departs from the spirit and scope of technical scheme of the embodiment of the present invention, should all cover the protection in the present invention Within the scope of.Therefore, protection scope of the present invention described should be defined by scope of the claims.

Claims (10)

  1. A kind of 1. safety check detection method, it is characterised in that including:
    The x-ray image of the X-ray machine collection in the screening machine that safety check terminal receives is obtained, the x-ray image is pre-processed, Obtain pretreated x-ray image;
    It is special according to the article of corresponding thing to be detected in pretreated x-ray image described in default deep learning model extraction Sign, the default deep learning model include the deep learning model based on convolutional neural networks;
    The article characteristics are identified using the grader based on the default deep learning model training, generation is corresponding The recognition result of the thing to be detected;
    The recognition result of the thing to be detected is sent to the safety check terminal, so that the safety check terminal shows the identification knot Fruit.
  2. 2. according to the method for claim 1, it is characterised in that it is described that the x-ray image is pre-processed, including:
    Smoothing denoising is carried out to the x-ray image of collection using neighborhood averaging, obtains the x-ray image after smoothing denoising;
    The marginal information of the image after the smoothing denoising is strengthened using histogram equalization method, obtains pretreated X Light image.
  3. 3. according to the method for claim 1, it is characterised in that the deep learning model based on convolutional neural networks is The article sample data for exceeding certain threshold value by quantity trains what is obtained, and the article sample data includes disobeying for different shape Picture corresponding to contraband goods.
  4. 4. according to the method for claim 1, it is characterised in that the training process of the grader includes:
    Utilize the depth characteristic of the deep learning model extraction article sample data based on convolutional neural networks;
    Based on machine learning algorithm, grader is trained to the depth characteristic;
    Wherein described article sample data include the X-ray picture for the different recognition results specified.
  5. A kind of 5. safety check detection means, it is characterised in that including:
    Pretreatment module, the x-ray image of the X-ray machine collection in the screening machine received for obtaining safety check terminal, to the X-ray Image is pre-processed, and obtains pretreated x-ray image;
    Characteristic extracting module, corresponding in the pretreated x-ray image according to default deep learning model extraction The article characteristics of thing to be detected, the default deep learning model include the deep learning model based on convolutional neural networks;
    As a result identification module, for utilizing the grader based on the default deep learning model training to the article characteristics It is identified, the recognition result of the corresponding thing to be detected of generation;
    Result display module, for the recognition result of the thing to be detected to be sent to the safety check terminal, so that the safety check Terminal shows the recognition result.
  6. 6. a kind of safety check detecting system, it is characterised in that including screening machine, safety check terminal and safety check identification equipment, the safety check X-ray machine is provided with the safety check case of machine, the safety check identification equipment includes safety check detection means as claimed in claim 5;Institute State X-ray machine, the safety check identification equipment is connected with the safety check terminal respectively;
    The X-ray machine, the x-ray image of the thing to be detected for gathering the security check passage by the screening machine, by the X-ray figure As sending to the safety check terminal;
    The safety check terminal, for when listening to the x-ray image of reception, the x-ray image being sent to safety check and identified Equipment;It is additionally operable to receive the recognition result for the thing to be detected that the safety check identification equipment is sent, the recognition result is led to Display screen is crossed to show.
  7. 7. system according to claim 6, it is characterised in that the recognition result includes two kinds of contraband and non-contraband Type;The system also includes warning device, and the warning device is connected with the safety check terminal;
    The safety check terminal, it is additionally operable to when the recognition result received is contraband, sends alarm signal to the report Alarm device, so that the warning device carries out alarm.
  8. 8. system according to claim 6, it is characterised in that the bottom of the security check passage of the screening machine is provided with pressure Sensor, the pressure sensor are connected with the safety check terminal;
    The pressure sensor, for gathering the pressure information born on the security check passage, by the pressure information send to The safety check terminal;
    The safety check terminal, it is additionally operable to be turned on or off the X-ray machine according to the pressure information.
  9. 9. according to the system described in claim any one of 6-8, it is characterised in that the safety check identification equipment includes Nvidia Jetson TX2 chips.
  10. 10. a kind of electronic equipment, including memory, processor, it is stored with and can runs on the processor on the memory Computer program, it is characterised in that the claims 1 to 4 are realized described in the computing device during computer program Method described in one.
CN201711126618.2A 2017-11-14 2017-11-14 Safety check detection method, device, system and electronic equipment CN107871122A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201711126618.2A CN107871122A (en) 2017-11-14 2017-11-14 Safety check detection method, device, system and electronic equipment

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201711126618.2A CN107871122A (en) 2017-11-14 2017-11-14 Safety check detection method, device, system and electronic equipment
PCT/CN2018/115498 WO2019096181A1 (en) 2017-11-14 2018-11-14 Detection method, apparatus and system for security inspection, and electronic device

Publications (1)

Publication Number Publication Date
CN107871122A true CN107871122A (en) 2018-04-03

Family

ID=61753983

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201711126618.2A CN107871122A (en) 2017-11-14 2017-11-14 Safety check detection method, device, system and electronic equipment

Country Status (2)

Country Link
CN (1) CN107871122A (en)
WO (1) WO2019096181A1 (en)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108519625A (en) * 2018-04-08 2018-09-11 安徽理工大学 Recognizable prohibited items safe examination system based on ray technology and deep learning
CN109001833A (en) * 2018-06-22 2018-12-14 天和防务技术(北京)有限公司 A kind of Terahertz hazardous material detection method based on deep learning
CN109086740A (en) * 2018-08-25 2018-12-25 李叶 Stagewise ferry safety check mechanism
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
CN109884721A (en) * 2018-12-10 2019-06-14 深圳极视角科技有限公司 Safety check prohibited items detection method, device and electronic equipment based on artificial intelligence
CN109946746A (en) * 2019-03-21 2019-06-28 长安大学 A kind of safe examination system and method based on deep neural network

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103308538A (en) * 2013-06-17 2013-09-18 开平市中铝实业有限公司 Security check system for bus
CN205210316U (en) * 2015-11-18 2016-05-04 上海英迈吉东影图像设备有限公司 Security check machine with energy -conserving function
CN106198580A (en) * 2016-08-26 2016-12-07 合肥丁点网络科技有限责任公司 A kind of X-ray screening machine luggage dangerous materials fast automatic detecting alarm device and method
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
US20170337467A1 (en) * 2016-05-18 2017-11-23 Nec Laboratories America, Inc. Security system using a convolutional neural network with pruned filters

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103744120A (en) * 2013-12-30 2014-04-23 中云智慧(北京)科技有限公司 Method of contrabands auxiliary identification and device thereof
CN106485268B (en) * 2016-09-27 2020-01-21 东软集团股份有限公司 Image identification method and device
CN106886054A (en) * 2017-04-13 2017-06-23 西安邮电大学 Dangerous material automatic identification equipment and method based on 3 D X-ray imaging
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

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103308538A (en) * 2013-06-17 2013-09-18 开平市中铝实业有限公司 Security check system for bus
CN205210316U (en) * 2015-11-18 2016-05-04 上海英迈吉东影图像设备有限公司 Security check machine with energy -conserving function
US20170337467A1 (en) * 2016-05-18 2017-11-23 Nec Laboratories America, Inc. Security system using a convolutional neural network with pruned filters
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
CN106198580A (en) * 2016-08-26 2016-12-07 合肥丁点网络科技有限责任公司 A kind of X-ray screening machine luggage dangerous materials fast automatic detecting alarm device and method

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
李莲 等: "基于卷积神经网络的辣椒识别", 《天津理工大学学报》 *

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
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
CN108519625A (en) * 2018-04-08 2018-09-11 安徽理工大学 Recognizable prohibited items safe examination system based on ray technology and deep learning
CN109001833A (en) * 2018-06-22 2018-12-14 天和防务技术(北京)有限公司 A kind of Terahertz hazardous material detection method based on deep learning
CN109086740A (en) * 2018-08-25 2018-12-25 李叶 Stagewise ferry safety check mechanism
CN109086740B (en) * 2018-08-25 2019-08-23 上海首安工程技术有限公司 Stagewise ferry safety check mechanism
CN109884721A (en) * 2018-12-10 2019-06-14 深圳极视角科技有限公司 Safety check prohibited items detection method, device and electronic equipment based on artificial intelligence
CN109946746A (en) * 2019-03-21 2019-06-28 长安大学 A kind of safe examination system and method based on deep neural network

Also Published As

Publication number Publication date
WO2019096181A1 (en) 2019-05-23

Similar Documents

Publication Publication Date Title
US8995780B2 (en) Compression of digital images of scanned documents
JP6547069B2 (en) Convolutional Neural Network with Subcategory Recognition Function for Object Detection
US8565482B2 (en) Local difference pattern based local background modeling for object detection
US8608073B2 (en) System and method for robust real-time 1D barcode detection
CN104166841B (en) The quick detection recognition methods of pedestrian or vehicle is specified in a kind of video surveillance network
CN105938559A (en) Digital image processing using convolutional neural networks
CN103984961B (en) A kind of image detecting method for being used to detect underbody foreign matter
Dougherty et al. Morphological texture-based maximum-likelihood pixel classification based on local granulometric moments
Abbasgholipour et al. Color image segmentation with genetic algorithm in a raisin sorting system based on machine vision in variable conditions
US20130279758A1 (en) Method and system for robust tilt adjustment and cropping of license plate images
US9008365B2 (en) Systems and methods for pedestrian detection in images
CN101350063B (en) Method and apparatus for locating human face characteristic point
EP3460765A1 (en) Banknote management method and system
Wang et al. On discrimination between photorealistic and photographic images
US20180114071A1 (en) Method for analysing media content
US9239948B2 (en) Feature descriptor for robust facial expression recognition
US20170124418A1 (en) System and a method for the detection of multiple number-plates of moving cars in a series of 2-d images
US20120033862A1 (en) Systems and methods for segmenting digital images
US9652851B2 (en) Side window detection in near-infrared images utilizing machine learning
US20140037159A1 (en) Apparatus and method for analyzing lesions in medical image
Khalifa et al. Malaysian Vehicle License Plate Recognition.
CN104091157A (en) Pedestrian detection method based on feature fusion
CN101599124B (en) Method and device for segmenting characters from video image
CN102163284A (en) Chinese environment-oriented complex scene text positioning method
CN104881770A (en) Express bill information identification system and express bill information identification method

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
TA01 Transfer of patent application right
TA01 Transfer of patent application right

Effective date of registration: 20191104

Address after: 518081 Yantian Modern Industrial Service Center, 3018 Shayan Road, Shatoujiao Street, Yantian District, Shenzhen City, Guangdong Province, 33 floors

Applicant after: SHENZHEN MALONG TECHNOLOGY CO., LTD.

Applicant after: Shenzhen Malong artificial intelligence research center

Address before: 518000, Shenzhen, Yantian District, Guangdong Province Industrial Street East International Creative port 2, 5 B

Applicant before: SHENZHEN MALONG TECHNOLOGY CO., LTD.