X-ray inspection equipment

Abstract

To accurately specify the number of cargoes.SOLUTION: An X-ray inspection device 1 includes: an imaging unit 10 which images multiple cargoes 2 being the inspection objects by using X-rays and outputs X-ray images of the inspection objects; an image analysis unit 34 which analyzes the X-ray images of the inspection objects and obtains the distribution of luminance values of the X-ray images of the inspection objects; a feature amount extraction unit 36 which extracts a feature amount of the inspection object on the basis of the distribution of the luminance values; a storage unit 50 which stores a reference feature amount being a feature amount obtained from the distribution of the luminance values of the reference X-ray image obtained by imaging one cargo; and a counting unit 38 which calculates the number of cargoes 2 being the inspection objects on the basis of the feature amount of the inspection object and the reference feature amount.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] This disclosure relates to an X-ray inspection device. [Background technology]

[0002] At customs and other authorities, containers undergoing import and export inspections sometimes contain materials such as grains packed in flexible container bags. To inspect the cargo inside such containers, X-ray inspection equipment is used to capture X-ray images of the cargo through the container.

[0003] For example, Patent Documents 1 and 2 disclose the use of an X-ray inspection device to identify the type of goods contained inside a container or carton box, and to determine whether there are any goods that have not been declared to customs or any dangerous goods, without opening the container or carton box containing the goods. [Prior art documents] [Patent Documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2018-112551 [Patent Document 2] Japanese Patent Application Publication No. 11-194102 [Overview of the Initiative] [Problems that the invention aims to solve]

[0005] However, the conventional X-ray inspection devices described in Patent Documents 1 and 2 above focus on identifying the type of cargo and the presence or absence of hazardous materials within a container, making it difficult to accurately determine the number of multiple cargo items within a container. In particular, when a large number of identical cargo items are stored stacked inside a container, it is difficult to accurately determine the number of these cargo items from the X-ray image in which the numerous cargo items overlap.

[0006] For this reason, customs inspectors have had no choice but to visually count the number of cargoes by observing X-ray images in which a plurality of cargoes overlap with each other. Accordingly, as the size of containers increases and the number of cargoes to be counted visually increases, the burden on inspectors has increased.

[0007] Accordingly, an object of the present disclosure is to provide an X-ray inspection apparatus capable of accurately specifying the number of cargoes. [Means for Solving the Problem]

[0008] In order to solve the above problem, the X-ray inspection apparatus of the present disclosure comprises: an imaging unit configured to image a plurality of cargoes as an inspection object using X-rays and output an X-ray image of the inspection object; an image analysis unit configured to analyze the X-ray image of the inspection object to obtain a luminance value distribution of the X-ray image of the inspection object; a feature quantity extraction unit configured to extract a feature quantity of the inspection object based on the luminance value distribution; a storage unit configured to store a reference feature quantity, which is a feature quantity obtained from a luminance value distribution of a reference X-ray image obtained by imaging one cargo; a counting unit configured to calculate the number of the cargoes in the inspection object based on the feature quantity of the inspection object and the reference feature quantity; picture, The items to be inspected are those in which the aforementioned multiple cargoes are stacked in one or more layers. The X-ray image of the object to be inspected is a grayscale image and includes at least one image region with varying shades depending on the number of layers in which the cargo is stacked. The image analysis unit generates a histogram representing the distribution of the brightness values ​​by analyzing the X-ray image of the object to be inspected. The histogram has at least one peak corresponding to the image region, where the intensity differs depending on the number of steps. The feature extraction unit extracts the feature of the subject to be examined for each peak in the histogram.

[0010] The feature quantity may include an average luminance value of an image region corresponding to each peak of the histogram and an area of the image region.

[0011] The counting unit is configured to: calculate the number of stacked layers of the cargoes based on the average luminance value that is the feature quantity of the inspection object and the average luminance value that is the reference feature quantity, ​The number of items of the goods to be inspected may be calculated based on the calculated number of layers, the area which is a characteristic of the item to be inspected, and the area which is a standard characteristic.

[0012] The system may further include an output image generation unit that generates an output image by superimposing at least one of the following onto the X-ray image of the object to be inspected: information regarding the number of items of the object to be inspected calculated by the counting unit, and information regarding the number of layers in which the items of the object to be inspected are stacked.

[0013] The aforementioned goods may be similar materials packaged in containers. [Effects of the Invention]

[0014] According to this disclosure, the number of cargo items can be accurately determined. [Brief explanation of the drawing]

[0015] [Figure 1] Figure 1 is a schematic diagram showing an X-ray inspection apparatus according to the first embodiment of the present disclosure. [Figure 2] Figure 2 is a block diagram showing the configuration of the image processing apparatus according to the same embodiment. [Figure 3] Figure 3 is an explanatory diagram showing the processing of an X-ray image of an object to be inspected by the image processing apparatus according to the same embodiment. [Figure 4] Figure 4 is an explanatory diagram showing the process of extracting reference features from a reference X-ray image using the image processing apparatus according to the same embodiment. [Figure 5] Figure 5 is an explanatory diagram showing the process of extracting feature quantities of an object to be inspected from an X-ray image of the object to be inspected using the image processing device according to the same embodiment. [Figure 6] Figure 6 is a flowchart showing the X-ray inspection method according to the same embodiment. [Figure 7] Figure 7 is an explanatory diagram showing a specific example of the output image according to the same embodiment. [Modes for carrying out the invention]

[0016] Embodiments of this disclosure will be described below with reference to the attached drawings. The dimensions, materials, and other specific numerical values ​​shown in the embodiments are merely examples for the purpose of facilitating understanding and do not limit this disclosure unless otherwise specified. In this specification and drawings, elements having substantially the same function or configuration are denoted by the same reference numerals to avoid redundant explanations, and elements not directly related to this disclosure are omitted from the illustrations.

[0017] [1. Overall configuration of the X-ray inspection system] First, with reference to Figure 1, the overall configuration of the X-ray inspection apparatus 1 according to the first embodiment of this disclosure will be described. Figure 1 is a schematic diagram showing the X-ray inspection apparatus 1 according to this embodiment.

[0018] As shown in Figure 1, the X-ray inspection apparatus 1 according to this embodiment is a device for capturing an X-ray image of the object to be inspected and inspecting the contents of the object based on the X-ray image. The X-ray inspection apparatus 1 may be, for example, a large-scale X-ray inspection apparatus installed at customs offices, etc., for inspecting goods that are subject to import and export inspections. However, the X-ray inspection apparatus 1 is not limited to this example, and can be applied to various other types of inspection devices as long as they are devices that use X-ray images to inspect the object to be inspected.

[0019] The X-ray inspection apparatus 1 comprises an X-ray imaging device 10 and an image processing device 20. The X-ray imaging device 10 and the image processing device 20 are connected to each other so as to be able to communicate with each other via a wired or wireless network or wiring.

[0020] The X-ray imaging device 10 is an example of the imaging unit of this disclosure. The X-ray imaging device 10 images the object to be inspected using X-rays 5. The X-ray imaging device 10 images a plurality of cargo 2 that are the object to be inspected using X-rays 5 and generates an X-ray image showing the cargo 2 that are the object to be inspected. The X-ray image may be a grayscale image that represents the object to be imaged (e.g., cargo 2) using shades of black and white. The X-ray imaging device 10 outputs the X-ray image of the object to be inspected obtained by the imaging process to the image processing device 20.

[0021] The image processing device 20 is an information processing device that performs various image processing and various calculation processing on X-ray images. Based on the X-ray image of the item to be inspected acquired from the X-ray imaging device 10, the image processing device 20 identifies the contents and quantity of the goods 2 to be inspected.

[0022] First, the image processing device 20 analyzes the X-ray image of the object to be inspected to determine the distribution of brightness values ​​in the X-ray image. The distribution of brightness values ​​is the distribution of brightness values ​​(pixel values) of multiple pixels contained in the X-ray image. The distribution of brightness values ​​may be, for example, a brightness value histogram. A brightness value histogram is a graph that represents the frequency distribution of brightness values ​​of multiple pixels contained in the X-ray image. The image processing device 20 generates a brightness value histogram based on the brightness values ​​of multiple pixels in the X-ray image of the object to be inspected.

[0023] Next, the image processing device 20 analyzes the histogram to classify (cluster) the distribution of brightness values ​​into one or more peaks. The image processing device 20 extracts feature quantities of the object to be inspected for each peak in the distribution of brightness values. The feature quantities of the object to be inspected are feature quantities related to the peaks in the distribution of brightness values ​​of the X-ray image of the object to be inspected.

[0024] Subsequently, the image processing device 20 compares the extracted feature quantities of the inspection target with pre-registered reference feature quantities and calculates the number of inspection target goods 2 based on the comparison result. Furthermore, the image processing device 20 overlays the information regarding the calculated number of inspection target goods 2 onto the X-ray image of the inspection target and displays it on the display unit.

[0025] The X-ray inspection device 1, equipped with the X-ray imaging device 10 and image processing device 20 described above, can automatically and accurately identify the number of cargo items 2 to be inspected contained in the container 3. Therefore, users (for example, customs inspectors) can easily determine the exact number of cargo items 2 in the container 3 to be inspected using the X-ray inspection device 1. Thus, users do not need to count the cargo items 2 by visually observing the X-ray images, significantly reducing the burden on the user's inspection work.

[0026] [2. Regarding cargo] Next, with reference to Figure 1, the cargo 2 that is the object of inspection by the X-ray inspection apparatus 1 according to this embodiment will be described.

[0027] As shown in Figure 1, cargo 2 is the object of inspection by the X-ray inspection device 1. In this embodiment, cargo 2 is housed inside a transport container 3. Each individual cargo 2 may be material packaged in a container such as a flexible container bag.

[0028] Here, the material may be, for example, grains such as soybeans, wheat, and corn, or other foodstuffs. The material may also be agricultural or livestock materials such as fertilizers and animal feed. Furthermore, the material may be building materials such as cement, ore, and limestone. Furthermore, the material may be industrial materials such as various metal materials, plastic materials, and liquid materials.

[0029] In this embodiment, an example is described in which cargo 2 is a particulate material such as grain packed in a container such as a flexible container bag. A flexible container bag is a bag made of synthetic fibers such as polyethylene or polypropylene, and is used for packing and transporting various materials such as grain. By packing and transporting materials in flexible container bags of similar size, the handling and transport of particulate materials becomes easier, and the amount of particulate materials can be easily managed by the number of flexible container bags. In addition, by stacking multiple flexible container bags packed with materials in multiple layers inside a container 3, storage space can be reduced.

[0030] However, the container for holding the materials of cargo 2 is not limited to the example of a flexible container bag. The container may be any container, such as other plastic containers or cardboard boxes, as long as it is made of a material that can transmit X-rays 5. If the container for cargo 2 is a flexible container bag, it becomes possible to stably stack multiple cargo 2s. However, a container that is not flexible may also be used.

[0031] Furthermore, as shown in Figure 1, the object to be inspected according to this embodiment is, for example, a container 3 in which multiple cargo items 2 are stacked in one or more layers. In the example in Figure 1, nine cargo items 2 are stacked in three layers, and the overall arrangement is a regularly stacked pyramidal shape. It is preferable that the multiple cargo items 2 to be inspected are stacked in a regularly structured manner. This allows the image processing device 20, described later, to more accurately estimate the number of layers of stacked cargo items 2 and the number of cargo items 2 in each layer. Therefore, when the image processing device 20, described later, calculates the number of cargo items 2 based on the analysis results of the X-ray image, it becomes possible to calculate a more accurate number. However, the method of stacking the cargo items 2 is not limited to the pyramidal stacking method shown in Figure 1; other regular stacking methods or random stacking methods may also be used.

[0032] Furthermore, it is preferable that the multiple cargo items 2 to be inspected are of the same type, and in particular, that they are the same type of material (such as soybeans) packed in containers of similar size (such as flexible container bags). This ensures that the X-ray images of each cargo item 2 are similar. Therefore, when the image processing device 20 described later calculates the number of cargo items 2 based on the analysis results of the X-ray images, it becomes possible to calculate a more accurate number.

[0033] Furthermore, cargo 2 does not have to be particulate material like the grains mentioned above, as long as it is an item to be transported. For example, cargo 2 may be various items such as products, parts, materials, machinery and equipment, or food products. Also, cargo 2 is not limited to materials packaged in a container, but may be an item not packaged in a container. In addition, although the cargo 2 to be inspected in this embodiment is stored in a container 3, the example is not limited to this. Cargo 2 may be stored exposed without being placed in a transport container such as a container 3.

[0034] [3. About X-ray imaging equipment] Next, with reference to Figure 1, the configuration of the X-ray imaging apparatus 10 according to this embodiment will be described.

[0035] As shown in Figure 1, the X-ray imaging device 10 (corresponding to the "imaging unit") is a device for imaging the cargo 2 contained in the container 3 and generating an X-ray image of the cargo 2. The X-ray imaging device 10 comprises an X-ray irradiation unit 12, an X-ray detection unit 14, an X-ray image generation unit 16, a controller 18, and a transport device (not shown) for transporting the container 3.

[0036] The X-ray irradiation unit 12 and the X-ray detection unit 14 are arranged opposite each other with a predetermined distance between them. In the example shown in Figure 1, the X-ray irradiation unit 12 and the X-ray detection unit 14 are arranged opposite each other in the vertical direction. The X-ray irradiation unit 12 includes an X-ray source that generates X-rays 5 and an X-ray irradiation device. The X-ray irradiation unit 12 irradiates the X-rays 5 toward the X-ray detection unit 14 in a predetermined direction (for example, the Z direction in Figure 1). The X-rays 5 pass through the object to be imaged and head toward the X-ray detection unit 14. The X-ray detection unit 14 is composed of, for example, a line sensor for X-ray detection. The X-ray detection unit 14 detects the X-rays 5 that have been irradiated from the X-ray irradiation unit 12 and have passed through the object to be imaged. The X-ray detection unit 14 outputs an imaging signal representing the detected X-rays 5.

[0037] The X-ray image generation unit 16 generates an X-ray image of the object to be imaged based on the imaging signal of the X-rays 5 detected by the X-ray detection unit 14. The X-ray image generation unit 16 includes, for example, an AD conversion element and an image memory. The AD conversion element performs AD conversion on the imaging signal output from the X-ray detection unit 14. The image memory stores the AD-converted imaging signal pixel by pixel.

[0038] The transport device consists of a belt conveyor and the like, and transports the container 3 containing the cargo 2 in a predetermined direction (for example, the X direction in Figure 1). The controller 18 controls each part of the X-ray imaging device 10, including the X-ray irradiation unit 12, the X-ray detection unit 14, the X-ray image generation unit 16, and the transport device.

[0039] With the X-ray imaging device 10 configured in this way, X-rays 5 can be scanned over the cargo 2 to be inspected, and an X-ray image of the cargo 2 can be acquired. The X-ray imaging operation by such an X-ray imaging device 10 will be described below.

[0040] The X-ray imaging device 10 operates the transport device with the container 3 placed on it. As a result, the container 3 moves relative to the X-ray irradiation unit 12 and the X-ray detection unit 14 in the scanning direction of the X-rays 5 (the X direction in Figure 1), passing between the X-ray irradiation unit 12 and the X-ray detection unit 14. At this time, the X-ray irradiation unit 12 irradiates the container 3, which is the object to be imaged, with X-rays 5. The X-ray detection unit 14 detects the X-rays 5 that have passed through the container 3 and the cargo 2 inside it from the X-ray irradiation unit 12. The X-ray image generation unit 16 generates an X-ray image showing multiple cargo items 2 based on the X-rays 5 detected by the X-ray detection unit 14.

[0041] In this embodiment, an X-ray image is generated of multiple cargo items 2 stacked inside the container 3, captured from above. This X-ray image is a two-dimensional image of the multiple cargo items 2 projected onto the XY plane, and corresponds to an image of the multiple cargo items 2 viewed through from above. However, the imaging direction of the X-ray image is not limited to the vertical direction. For example, the X-ray image may be an image of the multiple cargo items 2 captured from below (Z direction). Alternatively, the X-ray image may be a side view (Y direction) of the multiple cargo items 2 captured from the right or left side. Alternatively, the X-ray image may be a front view (X direction) of the multiple cargo items 2 captured from the front or back. Furthermore, multiple X-ray images of the cargo items 2 captured from multiple directions may be generated.

[0042] Furthermore, the X-ray imaging device 10 generates X-ray images of multiple cargo items 2 to be inspected (see Figure 5A, described later) in the manner described above. The X-ray imaging device 10 outputs the generated X-ray images to the image processing device 20.

[0043] Furthermore, the X-ray imaging device 10 can also generate a reference X-ray image (see Figure 4A, described later) of a single cargo 2 (reference cargo). The reference X-ray image is a reference X-ray image used to obtain different reference features depending on the type of cargo 2. Details of these reference X-ray images and reference features will be described later. The X-ray imaging device 10 captures reference X-ray images of multiple types of cargo 2 and outputs them to the image processing device 20. For example, the X-ray imaging device 10 captures a single flexible container bag containing soybeans as cargo 2 to generate a reference X-ray image related to soybeans. When capturing the reference X-ray image, it is preferable to capture a single cargo 2 that is not contained in a container 3. This makes it possible to obtain a reference X-ray image that accurately reflects the characteristics of the cargo 2.

[0044] Here, we will describe an example of the specifications of an X-ray image captured by the X-ray imaging apparatus 10 according to this embodiment.

[0045] The X-ray image according to this embodiment is, for example, a grayscale image that represents the imaged object being inspected using shades of black and white. A grayscale image does not contain color information, but only brightness information. The grayscale image may be, for example, an image in which the brightness value (pixel value) of one pixel is represented by 8 bits. The gradation of this 8-bit grayscale image is 256 levels (brightness value: 0 to 255). A brightness value of 0 represents black, and a brightness value of 255 represents white. By using a typical 256-level grayscale image, the characteristic image processing of this embodiment can be suitably realized. However, the gradation of the X-ray image may be other than 256 levels.

[0046] [4. Configuration of the image processing device] Next, the configuration of the image processing apparatus 20 according to this embodiment will be described with reference to Figure 2. Figure 2 is a block diagram showing the configuration of the image processing apparatus 20 according to this embodiment.

[0047] [4.1. Hardware configuration of the image processing unit] First, the hardware configuration of the image processing device 20 according to this embodiment will be described. Figure 2 also shows an example of the hardware configuration of the image processing device 20 according to this embodiment.

[0048] As shown in Figure 2, the image processing device 20 includes a processor 21, memory 22, storage 23, communication device 24, input device 25, output device 26, and bus 27.

[0049] The processor 21 is an arithmetic processing unit installed in a computer. The processor 21 may consist of, for example, a CPU (Central Processing Unit), but it may also consist of other microprocessors. Furthermore, the processor 21 may consist of one or more processors. The processor 21 executes programs stored in the memory 22 or other storage media. This enables the execution of various processes in the image processing device 20.

[0050] The program is a computer program that includes instructions to be executed by a computer. The program may be provided to the image processing device 20, for example, by distribution from an external device via a communication network. Alternatively, the program may be provided to the image processing device 20 via a non-transitory computer-readable medium. By installing the program in the image processing device 20, the image processing device 20 becomes capable of realizing various functions defined by the program.

[0051] Memory 22 is a storage medium that stores programs and various other data. Memory 22 includes, for example, RAM (Random Access Memory) and ROM (Read Only Memory). ROM is a non-volatile memory that stores programs used by the processor 21 and data necessary to run those programs. RAM is a volatile memory that temporarily stores data such as variables, arithmetic parameters, and calculation results used in processes executed by the processor 21. Programs stored in ROM are read into RAM and executed by the processor 21, such as the CPU.

[0052] Storage 23 is a storage device for storing various types of information and data. Storage 23 includes, for example, a recording medium such as semiconductor memory, a hard disk, or an optical disc, and a drive for reading data from or writing data to the recording medium. Storage 23 can store a larger amount of data than memory 22. Storage 23 may be internal storage built into the image processing device 20, or it may be external storage connected via the external input / output terminals of the image processing device 20. Alternatively, storage 23 may be online storage connected via a communication network.

[0053] The communication device 24 is a device for communicating with an external device connected to the image processing device 20 by wire or wireless connection. The communication device 24 establishes a communication connection with the external device according to a predetermined protocol and transmits and receives various information and data with the external device.

[0054] The input device 25 is a device used by the user to input information into the image processing device 20. The input device 25 includes, for example, a touch sensor, keyboard, keypad, mouse, remote controller, button, switch, or dial. The input device 25 may also include an input device for voice input, such as a microphone or a voice recognition module. The input device 25 may also include a remote control module that receives user input from a remote device for remotely operating the image processing device 20. When the input device 25 receives an input operation from the user, it transmits an input signal corresponding to that input operation to the processor 21.

[0055] The output device 26 is a device for outputting information and data to the outside of the image processing device 20. The output device 26 includes a display device that displays information such as text, graphics, and images, and an audio output device that outputs sound. The display device comprises a display screen and an image display module. The display may be, for example, a liquid crystal display (LCD), a plasma display (PDP), an organic light-emitting diode (OLED), or a cathode ray tube (CRT). The display device may also be a touch panel with a touch sensor on the display screen. The audio output device comprises a speaker and an audio output module.

[0056] Bus 27 interconnects the processor 21, memory 22, storage 23, communication device 24, input device 25, and output device 26. This allows various types of information and data to be transmitted and received between these devices.

[0057] [4.2. Functional Configuration of Image Processing Devices] Next, the functional configuration of the image processing apparatus 20 according to this embodiment will be described with reference to Figures 2 and 3. Figure 3 is an explanatory diagram showing the processing of an X-ray image of an object to be inspected by the image processing apparatus 20 according to this embodiment.

[0058] As shown in Figure 2, the image processing device 20 comprises a database creation unit 30, a preprocessing unit 32, an image analysis unit 34, a feature extraction unit 36, a counting unit 38, an output image generation unit 40, and a cargo database 50. The processor 21 of the image processing device 20 performs calculations based on a program, thereby realizing each of these functional units of the image processing device 20.

[0059] (1) Database creation unit 30 and cargo database 50 The database creation unit 30 (hereinafter referred to as "DB creation unit 30") creates and manages the cargo database 50 (hereinafter referred to as "cargo DB 50").

[0060] Cargo DB50 is an example of a storage unit that stores the reference features of this disclosure. Cargo DB50 is a database that stores various information about various types of cargo 2. Information about each cargo 2 includes, for example, a file of a reference X-ray image of cargo 2, reference features of cargo 2, identification information representing the type of cargo 2, and information representing the number and size of cargo 2. The reference features of cargo 2 are features obtained from the distribution of brightness values ​​in a reference X-ray image of one cargo 2. By comparing the reference features of this one cargo 2 with the features of multiple cargo 2 to be inspected, it becomes possible to identify the type and number of multiple cargo 2 to be inspected.

[0061] The DB creation unit 30 registers information about various types of cargo (reference cargo) in the cargo DB 50. For example, let's consider the case where information about soybeans packed in flexible container bags is registered in the cargo DB 50 as a new type of cargo. In this case, first, the X-ray imaging device 10 takes an X-ray image of the soybeans packed in one flexible container bag and generates an X-ray image of the soybeans (reference X-ray image: see Figure 4A). The X-ray imaging device 10 outputs the reference X-ray image of the soybeans to the image processing device 20. Next, the DB creation unit 30 of the image processing device 20 acquires the reference X-ray image of the soybeans from the X-ray imaging device 10. Then, the image analysis unit 34, which will be described later, analyzes the distribution of brightness values ​​in the reference X-ray image of the soybeans. Furthermore, the feature extraction unit 36, which will be described later, extracts features related to the distribution of brightness values ​​in the reference X-ray image of the soybeans as reference features of the soybeans. Subsequently, the DB creation unit 30 associates the standard feature quantities of the soybean, the standard X-ray image file of the soybean, the identification information representing the soybean, and information representing the size of one flexible container bag, and registers them in the cargo DB 50.

[0062] In this way, for each type of cargo 2, a reference X-ray image of one cargo (reference cargo) is acquired in advance, and reference features are determined. Then, the reference X-ray image and reference features of the cargo 2 are registered in advance in the cargo DB 50. As a result, when the image processing device 20 acquires an X-ray image of the cargo 2 to be inspected, it can use the reference features of various cargo 2 registered in the cargo DB 50 to identify the type and number of cargo 2 to be inspected.

[0063] (2) Preprocessing unit 32 When the image processing device 20 acquires an X-ray image of the cargo 2 to be inspected from the X-ray imaging device 10 (see Figure 3A), the preprocessing unit 32 first performs preprocessing on the X-ray image. This preprocessing may include, for example, image cropping, image size adjustment, and brightness value adjustment. For example, the preprocessing unit 32 may crop the image region containing the cargo 2 from the acquired X-ray image of the cargo to be inspected. The preprocessing unit 32 may also adjust the size and brightness values ​​to be suitable for extracting features from the X-ray image.

[0064] Furthermore, the X-ray image of the object to be inspected according to this embodiment is an X-ray image of multiple cargo 2 contained in the container 3, taken from outside the container 3. Therefore, the X-ray image is affected not only by the cargo 2 to be inspected but also by the container 3. Accordingly, the preprocessing unit 32 may perform image processing on the X-ray image acquired from the X-ray imaging device 10 to eliminate the influence of the container 3. This makes it possible to more accurately identify the documents and quantities of cargo 2 shown in the X-ray image.

[0065] (3) Image analysis unit 34 The image analysis unit 34 obtains the distribution of brightness values ​​of the X-ray image of the object to be inspected by analyzing the pre-processed X-ray image of the object to be inspected. Here, the distribution of brightness values ​​is the distribution of brightness values ​​(pixel values) of multiple pixels contained in the X-ray image of the object to be inspected. Specifically, as shown in Figure 3B, the image analysis unit 34 generates a histogram (brightness value histogram) that represents the distribution of brightness values ​​of the X-ray image of the object to be inspected. The histogram is a graph of the distribution of brightness values ​​of multiple pixels in the X-ray image.

[0066] For example, if the X-ray image is a 256-level grayscale image as described above, the brightness value (pixel value) of each pixel in the X-ray image has an 8-bit value. Therefore, the horizontal axis of the histogram created from such an X-ray image represents the brightness value (0 to 255), and the vertical axis represents the number of pixels (frequency) that have each brightness value. By generating a histogram of an X-ray image, it is possible to understand the distribution of brightness values ​​of multiple pixels in the X-ray image. The distribution of brightness values ​​corresponds to the intensity of the image region representing cargo 2 in the X-ray image.

[0067] (4) Feature extraction unit 36 The feature extraction unit 36 ​​extracts features of the inspection target based on the distribution of brightness values ​​obtained by the image analysis unit 34. Here, the features of the inspection target are features related to the distribution of brightness values ​​in the X-ray image of the inspection target cargo 2. For example, the feature extraction unit 36 ​​extracts features of the inspection target based on a histogram representing the distribution of brightness values ​​in the X-ray image of the inspection target cargo 2.

[0068] Specifically, as shown in Figures 3B and 3C, the histogram has at least one peak depending on the number of layers in which the multiple items 2 being inspected are stacked. In the example in Figure 3C, the histogram has two peaks, P1 and P2. These peaks P1 and P2 appear as many times as there are layers in which the multiple items 2 being inspected are stacked.

[0069] For example, consider an X-ray image of stacked cargo 2 taken from above, where there are areas where cargo 2 is stacked one layer and areas where cargo 2 is stacked two layers high. In this case, as shown in Figure 3A, the density of the grayscale image differs between the first image region representing the area where cargo 2 is stacked one layer high (the lighter area in the X-ray image in Figure 3A) and the second image region representing the area where cargo 2 is stacked two layers high (the darker area in the X-ray image in Figure 3A). Thus, the X-ray image to be inspected includes at least one image region where the density differs depending on the number of layers of cargo 2 stacked.

[0070] The density of these image regions appears as peaks in the histogram. Therefore, in the histograms of the examples in Figures 3B and 3C, due to the difference in density between the first and second image regions, there is one peak P1 corresponding to the first image region (lighter region) and another peak P2 corresponding to the second image region (darker region).

[0071] Therefore, the feature extraction unit 36 ​​extracts the features of the subject to inspection for each of the histogram peaks P1 and P2. For example, as shown in Figure 3C, the feature extraction unit 36 ​​derives the normal distribution peaks P1' and P2' corresponding to the histogram peaks P1 and P2, respectively. Then, based on the average brightness values ​​L1 and L2 and a predetermined deviation σ of the normal distribution peaks P1' and P2', the feature extraction unit 36 ​​calculates the features of the subject to inspection for each of the histogram peaks P1 and P2. This yields the features related to peak P1 corresponding to the first image region (light region) where cargo 2 is stacked in one layer, and the features related to peak P2 corresponding to the second image region (dark region) where cargo 2 is stacked in two layers.

[0072] As described above, the feature extraction unit 36 ​​extracts the features of the object to be inspected by analyzing the brightness value histogram generated by the image analysis unit 34. Details of the feature extraction process by the feature extraction unit 36 ​​will be described later.

[0073] (5) Counting unit 38 The counting unit 38 calculates the number of items of the cargo 2 to be inspected based on the features of the cargo 2 to be inspected extracted by the feature extraction unit 36 ​​and the reference features pre-registered in the cargo database 50. For example, when the counting unit 38 receives the features of the cargo 2 to be inspected from the feature extraction unit 36, it reads the reference features corresponding to the cargo 2 to be inspected from the cargo database 50. Then, the counting unit 38 performs calculation processing using a predetermined algorithm for counting using the features of the cargo 2 to be inspected and the reference features of the cargo 2 to be inspected. In this way, the counting unit 38 derives the number of multiple cargo 2 to be inspected from the features of multiple cargo 2 to be inspected and the reference features of the cargo 2 to be inspected. Details of the calculation process of the number of cargo 2 by the counting unit 386 will be described later.

[0074] (6) Output image generation unit 40 The output image generation unit 40 generates an output image representing the number of cargo 2 calculated by the counting unit 38. Preferably, the output image is one in which at least one of the following is superimposed on the X-ray image of the cargo to be inspected: information regarding the number of cargo 2 to be inspected and information regarding the number of layers in which the cargo 2 to be inspected is stacked. The output image generation unit 40 generates an output image (see Figure 7) in which this information is superimposed on the X-ray image. The output image generated by the output image generation unit 40 is displayed on the display unit (display, etc.) of the output device 26 of the image processing device 20.

[0075] [5. Feature extraction process] Next, with reference to Figures 4 and 5, the feature extraction process by the image processing device 20 according to this embodiment will be described in detail. Figure 4 is an explanatory diagram showing the process of extracting reference features from a reference X-ray image by the image processing device 20 according to this embodiment. Figure 5 is an explanatory diagram showing the process of extracting features of an object to be inspected from an X-ray image of the object to be inspected by the image processing device 20 according to this embodiment.

[0076] First, referring to Figure 4, the process of extracting reference features from a reference X-ray image by the feature extraction unit 36 ​​of the image processing apparatus 20 according to this embodiment will be described.

[0077] As shown in Figure 4A, the reference X-ray image is an X-ray image of one cargo 2 (reference cargo). Therefore, naturally, the number of stacking layers N of the one cargo 2 shown in the reference X-ray image is... ref is "1" (N ref =1). Therefore, in the grayscale image of the reference X-ray image, image region A represents the single cargo 2 that is captured. ref The brightness value is generally constant (for example, a brightness value of around 150-170). Therefore, as shown in Figure 4B, the histogram generated from the reference X-ray image is in the image region A. ref One peak P corresponding to ref It has this peak P. refThe luminance value represented by corresponds to the luminance values of a plurality of pixels included in image area A in the reference X-ray image ref (an area in which one cargo 2 is captured).

[0078] The feature extraction unit 36 extracts a reference feature quantity from the histogram of the reference X-ray image of the one cargo 2. First, as shown in FIG. 4C, the feature extraction unit 36 clusters (approximates) one peak P of the histogram of the reference X-ray image ref with a single normal distribution. The broken-line peak P ref ' is the peak of the approximated normal distribution. As the clustering method, a known clustering method can be used, for example, a method based on mean shift or a Gaussian mixture model.

[0079] Next, the feature extraction unit 36 obtains the average luminance value L of the normal distribution of peak P ref '. Here, the average luminance value L ref is obtained. Here, the average luminance value L ref is, for example, the luminance value at the vertex of the peak P ref ' of the normal distribution. Further, the feature extraction unit 36 sets a predetermined deviation σ (e.g., standard deviation) for the normal distribution of peak P ref '.

[0080] Next, the feature extraction unit 36 obtains the area S (number of pixels) of the histogram within the range of the average luminance value L ref ± the predetermined deviation σ. Here, the area S ref is the number of pixels having luminance values within the range of the average luminance value L ref ± the predetermined deviation σ. The area S ref corresponds to the area (number of pixels) of image area A ref where one cargo 2 is captured in the reference X-ray image shown in FIG. 4A ref Then, the feature extraction unit 36 outputs the average luminance value L ref and the area S ref calculated as described above as the reference feature quantity of one cargo 2 (reference cargo).

[0081] Next, referring to Figure 5, the process of extracting features of the subject to be inspected from the X-ray image of the subject to be inspected by the feature extraction unit 36 ​​of the image processing apparatus 20 according to this embodiment will be described.

[0082] As shown in Figure 5A, the X-ray image of the object to be inspected is an X-ray image of multiple cargo items 2 that are to be inspected. The multiple cargo items 2 may be stacked in multiple layers within the container 3 (the number of stacking layers N is 2 or more), or they may be laid flat in a single layer (the number of stacking layers N is 1). Therefore, the number of stacking layers N of the multiple cargo items 2 shown in the X-ray image of the object to be inspected is unknown. Note that the X-ray image of the object to be inspected shown in Figure 5A shows, for example, a state in which three cargo items 2 are stacked in two layers (two cargo items 2 are stacked in the first layer and one cargo item 2 is stacked in the second layer).

[0083] Therefore, in the grayscale X-ray image shown in Figure 5A, there are multiple image regions (first image regions A1, A1 and second image region A2) with mutually different densities (luminance values) depending on the number of stacking layers N of cargo 2. Specifically, there are two first image regions A1, A1 representing cargo 2 stacked one layer at a time, and one second image region A2 representing cargo 2 stacked two layers at a time.

[0084] The brightness values ​​in the first image region A1 are generally constant (for example, around 150-170). The brightness values ​​in the second image region A2 are also generally constant (for example, around 120-140). The brightness values ​​in the second image region A2 are lower than those in the first image region A1, resulting in a brightness value closer to black. In other words, the image density of the second image region A2 is higher than that of the first image region A1. This is because the transmittance of X-rays 5 is lower in the second image region A2, where cargo 2 is stacked in two layers, than in the first image region A1, where cargo 2 is stacked in one layer.

[0085] Therefore, as shown in Figure 5B, the histogram generated from the X-ray image of the object being inspected has multiple peaks P1 and P2 depending on the difference in the number of stacking layers N. In other words, the histogram has multiple peaks P1 and P2 depending on the multiple image regions A1 and A2 with different levels of density. Specifically, the histogram has one peak P1 corresponding to the first image regions A1, A1 and one peak P2 corresponding to the second image region A2. The brightness value represented by peak P1 corresponds to the brightness values ​​of multiple pixels contained in the two first image regions A1, A1 (regions showing one-layered cargo 2) in the X-ray image. Similarly, the brightness value represented by peak P2 corresponds to the brightness values ​​of multiple pixels contained in one second image region A2 (region showing two-layered cargo 2) in the X-ray image.

[0086] The feature extraction unit 36 ​​extracts features from the histogram of the X-ray image to be inspected, for each peak P1 and P2. First, as shown in Figure 5C, the feature extraction unit 36 ​​clusters (approximates and classifies) the multiple peaks P1 and P2 of the histogram of the X-ray image to be inspected using a normal distribution. The dashed peak P1' shown in Figure 5C is the normal distribution peak that approximates the peak P1 on the right side of the histogram. Similarly, the dashed peak P2' is the normal distribution peak that approximates the peak P2 on the left side of the histogram. As for the clustering method, known clustering methods such as mean shift or methods based on a Gaussian mixture model can be used.

[0087] Next, the feature extraction unit 36 ​​extracts the features to be inspected for each of the peaks P1 and P2 of the histogram. Specifically, first, the feature extraction unit 36 ​​calculates the average brightness value L1 of the normal distribution of peak P1' corresponding to peak P1, and the average brightness value L2 of the normal distribution of peak P2' corresponding to peak P2. Here, the average brightness values ​​L1 and L2 are, for example, the brightness values ​​of the peaks P1' and P2' of the normal distribution, respectively. Furthermore, the feature extraction unit 36 ​​sets predetermined deviations σ1 and σ2 (for example, standard deviations) of the normal distributions of peaks P1' and P2', respectively.

[0088] Next, the feature extraction unit 36 ​​calculates the area S1 (number of pixels) of the histogram within the range of average brightness value L1 ± predetermined deviation σ1, and the area S2 (number of pixels) of the histogram within the range of average brightness value L2 ± predetermined deviation σ2. Here, areas S1 and S2 are the number of pixels having brightness values ​​within the range of average brightness values ​​L1, L2 ± predetermined deviations σ1 and σ2, respectively. Area S1 corresponds to the total area (number of pixels) of the two first image regions A1 and A1 in the X-ray image shown in Figure 5A, which show the single-layered cargo 2. Similarly, area S2 corresponds to the area (number of pixels) of one second image region A2 in the X-ray image shown in Figure 5A, which shows the double-layered cargo 2. The feature extraction unit 36 ​​then outputs the average brightness values ​​L1 and L2 and areas S1 and S2 calculated as described above as reference features for the multiple cargo 2 to be inspected.

[0089] As described above, the feature extraction unit 36 ​​according to this embodiment extracts features of the subject to be inspected by using information on the distribution of brightness values ​​obtained from the X-ray image of the subject to be inspected, without referring to separate training data. Specifically, the feature extraction unit 36 ​​approximates the brightness value histogram of the X-ray image of the subject to be inspected with a normal distribution, clusters multiple peaks, and extracts the features of each peak. Such feature extraction processing corresponds to processing using unsupervised learning.

[0090] However, the system is not limited to such examples. For instance, the feature extraction unit 36 ​​can also extract features from X-ray images of the object being inspected using a model obtained through supervised learning. For example, when extracting features from X-ray images of the object being inspected by repurposing a neural network used for object detection, prior information (such as X-ray images of each stacking state of cargo 2) is required to train the neural network. In this case, the feature extraction process corresponds to a process using supervised learning.

[0091] The above describes an example of feature extraction processing when multiple cargo items 2 are stacked in two layers, that is, when the number of stacking layers N is 2. However, this disclosure is not limited to such an example. For example, when multiple cargo items 2 are stacked in three or more layers, that is, when the number of stacking layers N is 3 or more, the feature extraction processing can be performed in the same manner as above. In this case, the X-ray image of the object to be inspected will contain three or more image regions A1, A2, A3, ... with different shades depending on the number of stacking layers, and the histogram will contain three or more peaks P1, P2, P3, ... Therefore, feature quantities corresponding to each of these three or more peaks P1, P2, P3, ... are extracted.

[0092] [6. Calculation process for the number of cargo items] Next, with reference to Figures 4 and 5, the process for calculating the number M of cargo 2 by the image processing device 20 according to this embodiment will be described in detail.

[0093] As described above, the feature quantities of the object to be inspected according to this embodiment include the average brightness values ​​L1 and L2 of each image region A1 and A2 corresponding to each peak P1 and P2 of the histogram shown in Figure 5, and the areas S1 and S2 (number of pixels) of each image region A1 and A2. The reference feature quantity is the peak P of the histogram shown in Figure 4. ref Image region A corresponding to this region ref Average brightness value L ref And the image region A ref Area S ref (Number of pixels) is included.

[0094] The counting unit 38 of the image processing device 20 first calculates the number of layers Nn of cargo 2 stacked in each image region An in the X-ray image. Then, the counting unit 38 calculates the number M of cargo 2 to be inspected using the number of layers Nn.

[0095] Specifically, first, the counting unit 38 uses the average brightness value Ln (Ln=L1,L2,···,Lm), which is a feature of the object to be inspected, and the average brightness value L, which is a reference feature. refBased on this, the number of layers Nn (Nn=N1,N2,...,Nm) of cargo 2 stacked in each image region An (An=A1,A2,...,Am) in the X-ray image is calculated. Next, the counting unit 38 calculates the calculated number of stacking layers Nn, the area Sn (Sn=S1,S2,...,Sm) which is a feature quantity of the object to be inspected, and the area S ref Based on this, the number M of cargo 2 to be inspected is calculated. The process for calculating the number M of cargo 2 is explained in detail below.

[0096] Due to the characteristics of X-ray images, the following relationships exist between the average brightness value Ln of each image region An in the X-ray image of an object under inspection with an unknown stacking layer N (i.e., the average brightness values ​​L1, L2, ..., Lm, which are characteristic quantities of the object under inspection) and the actual stacking layer Nn of each image region An (Nn = N1, N2, ..., Nm): (1) and (2). From these relationships (1) and (2), the stacking layer Nn of each image region An can be calculated inversely.

[0097] Ln = g × exp(-μ × Nn) ... (1) μ = -log(L) ref / g) ···(2) Ln: The average brightness value of each image region An in the X-ray image of the subject being examined (= average brightness value Ln of the feature quantities of the subject being examined) g: Number of grayscale levels in the X-ray image (e.g., g=255) Nn: Number of stacking layers of cargo 2 in each image region An of the X-ray image being inspected. μ: Mass absorption coefficient (per stage) L ref :Average brightness value of cargo 2 per tier (=Average brightness value L of the reference feature) ref ) n: An ordinal number representing each image region An in the X-ray image (n=1,2,···,m) m: Number of image regions An in the X-ray image (= Number of peaks Pn in the histogram) However, m is an integer greater than or equal to 1, and n is an integer greater than or equal to 1 and less than or equal to m.

[0098] In this embodiment, for example, the number of grayscale levels g in the X-ray image is 256, and the brightness value of each pixel in the X-ray image ranges from 0 (black) to 255 (white). It is assumed that the brightness value of all pixels in an X-ray image that shows nothing is 255.

[0099] Using equations (1) and (2) above, the number of stacking layers Nn of cargo 2 in each image region An can be calculated. Then, using the following equation (3), Nn, Sn, and S ref From this, the number of cargo 2 units Mn (Mn = M1, M2, ..., Mm) in each image region An can be calculated. Then, by calculating the sum of Mn using equation (4), the total number of cargo 2 units visible in the X-ray image can be determined.

[0100] Mn = Nn × Sn / S ref ...(3) M = M1 + M2 + ... + Mm ... (4) M: The total number of cargo 2 units in the X-ray image being inspected. Mn: The number of cargo 2 in each image region An of the X-ray image being inspected (Mn = M1, M2, ..., Mm) Sn: Area of ​​each image region An in the X-ray image of the subject being examined (Sn = S1, S2, ..., Sm) S ref : Area of ​​the reference feature for cargo 2 with 1 item per layer (= Area of ​​reference feature S) ref )

[0101] Here, as a specific example of the calculation process for the number of cargo items M by the counting unit 38 as described above, we will explain the case where three cargo items 2 are stacked in two layers, as shown in Figure 5. In this case, the number m of image regions A1 and A2 in the X-ray image of the object to be inspected shown in Figure 5A is 2 (m=2). Therefore, the ordinal numbers n=1,2.

[0102] First, the counting unit 38 calculates the number of cargo items M1 in the first image region A1 where the stacking height N1 is 1. Specifically, the counting unit 38 first calculates the average brightness value L1 of the first image region A1 and the average brightness value L of the reference feature using equations (1) and (2). refFrom this, the number of stacking layers N1 of the first image region A1 is calculated to be "1" (N1=1). Next, the counting unit 38 uses equation (3) to calculate the area S1 of the first image region A1 and the area S of the reference feature. ref From the ratio (=1), we can calculate that the number of cargo 2 items M1 in the first image region A1 is "1" (M1 = N1 × S1 / S ref (=1×1=1)

[0103] Next, the counting unit 38 calculates the number of cargo items M2 in the second image region A2 where the stacking height N2 is 2. Specifically, the counting unit 38 first calculates the average brightness value L2 of the second image region A2 and the average brightness value L of the reference feature using equations (1) and (2). ref From this, the number of stacking layers N2 in the second image region A2 is calculated to be "2" (N2=2). Next, the counting unit 38 uses equation (3) to calculate the area S2 of the second image region A2 and the area S of the reference feature. ref From the ratio (=1), we can calculate that the number of cargo items M2 in the second image region A2 is "2" (M2 = N2 × S2 / S ref (=2×1=1)

[0104] Subsequently, the counting unit 38 adds M1 and M2 calculated above using formula (4) to determine that the number of cargo items M is "3" (M=M1+M2=1+2=3).

[0105] As described above, the counting unit 38 of the image processing device 20 counts the average brightness value Ln and area Sn of each image region An, which are feature quantities of the object to be inspected, and the average brightness value L, which is a reference feature quantity. ref and area S ref From this, the number M of cargo 2 visible in the X-ray image can be calculated. As a result, the number M of cargo 2 to be inspected can be automatically and accurately calculated simply by loading the X-ray image of the cargo to be inspected into the image processing device 20.

[0106] [7. X-ray examination method] Next, with reference to Figure 6, the X-ray inspection method using the X-ray inspection apparatus 1 according to this embodiment will be described in detail. Figure 6 is a flowchart of the X-ray inspection method according to this embodiment.

[0107] As shown in Figure 6, prior to the inspection of cargo 2 (S40-S90), in S10-S30, a process is executed to determine standard features for each type of standard cargo and register them in advance in the cargo DB 50.

[0108] First, the X-ray imaging device 10 of the X-ray inspection device 1 images a reference cargo 2 (reference cargo) using X-rays 5 to generate a reference X-ray image (see Figure 4A) (S10). At this time, it is preferable that the cargo 2 (reference cargo) is imaged while not contained in the container 3. This makes it possible to generate a reference X-ray image that accurately reflects the characteristic quantities of the brightness value distribution of the reference cargo. The X-ray imaging device 10 outputs the captured reference X-ray image to the image processing device 20.

[0109] Next, the image processing device 20 analyzes the reference X-ray image captured in S10 and extracts reference features of the reference cargo (S20). Specifically, first, the preprocessing unit 32 preprocesses the reference X-ray image. Then, the image analysis unit 34 analyzes the reference X-ray image to generate a histogram representing the distribution of brightness values ​​of the reference X-ray image (see Figure 4B). Furthermore, the feature extraction unit 36 ​​extracts the peak P of the histogram of the reference X-ray image. ref Average brightness value L ref and area S ref These are extracted as reference features (see Figure 4C). The process for extracting these reference features is the same as the process for extracting the features of the data to be inspected (S50-S70), which will be described later, so a detailed explanation is omitted.

[0110] Subsequently, the DB creation unit 30 of the image processing device 20 associates the reference feature quantities extracted in S20 with the reference X-ray image file, the identification information of the reference cargo, and the information representing the size of the reference cargo, and registers them in the cargo DB 50 (S30).

[0111] In steps S10 to S30 above, reference X-ray images of various cargo 2 (reference cargo) are acquired in advance to extract reference features, and the reference X-ray images, reference features, and identification information of the cargo 2 are registered in the cargo DB 50 in advance. This makes it possible to identify the type and quantity M of the cargo 2 to be inspected in subsequent inspections of cargo 2 (S40 to S90) based on the reference features and identification information registered in the cargo DB 50 in advance.

[0112] Next, in steps S40-S90, the process of inspecting cargo 2 is executed.

[0113] First, the X-ray imaging device 10 images the multiple cargo items 2 to be inspected using X-rays 5 to generate an X-ray image of the cargo items 2 (S40). At this time, the cargo items 2 may be imaged while still contained in the container 3. This allows the cargo items 2 to be inspected without opening the container 3 and removing the cargo items 2. The X-ray imaging device 10 outputs the imaged X-ray image of the cargo items 2 to the image processing device 20.

[0114] Next, the preprocessing unit 32 of the image processing device 20 performs preprocessing on the reference X-ray image captured in S40 (S50). This preprocessing includes, for example, cutting out the portion of the X-ray image in which the cargo 2 is visible, adjusting the image size, adjusting the brightness value, and image processing to eliminate the influence of the container 3.

[0115] Furthermore, the image analysis unit 34 of the image processing device 20 analyzes the X-ray image of the object to be inspected, which was preprocessed in S50, to generate a histogram (see Figures 3B and 5B) representing the distribution of brightness values ​​of the X-ray image (S60).

[0116] Subsequently, the feature extraction unit 36 ​​of the image processing device 20 extracts features of the object to be inspected based on the histogram generated in S60 (S70). Specifically, the feature extraction unit 36 ​​clusters the histogram of the X-ray image to be inspected using a normal distribution to obtain each peak P1 and P2 in the histogram (see Figures 3C and 5C). Each peak P1 and P2 corresponds to image regions A1 and A2 in the X-ray image. The feature extraction unit 36 ​​then extracts the average brightness values ​​L1 and L2 and areas S1 and S2 of these peaks P1 and P2 as features of the object to be inspected. The details of this feature extraction process for the object to be inspected are as described above, so a detailed explanation is omitted.

[0117] Next, the counting unit 38 of the image processing device 20 calculates the number M of the cargo 2 to be inspected based on the feature quantities of the cargo to be inspected extracted in S70 and the reference feature quantities that have been pre-registered in the cargo DB 50 in S30 (S80). Specifically, the counting unit 38 calculates the number M of the cargo 2 to be inspected based on the feature quantities of the cargo to be inspected, which are the average brightness values ​​L1 and L2 and the areas S1 and S2, and the reference feature quantity, which is the average brightness value L ref and area S ref Using this method, the number M of the cargo 2 to be inspected is calculated by performing calculations according to a predetermined algorithm. The details of the process for calculating the number M of cargo 2 to be inspected are as described above, so a detailed explanation is omitted.

[0118] Subsequently, the output image generation unit 40 of the image processing device 20 generates an output image representing the number of cargo 2 calculated in S80 and displays it on the display unit (S90). The output image is, for example, an image in which the number M of cargo 2 to be inspected and the number of stacking layers N of cargo 2 are superimposed on the X-ray image.

[0119] Figure 7 is an explanatory diagram showing a specific example of an output image according to this embodiment. As shown in Figure 7, in the output image, among the X-ray images of the object to be inspected, image regions An where the number of stacking layers Nn of cargo 2 is the same are surrounded by frames Fn of the same color. For example, two first image regions A1, A1 where the number of stacking layers N1 is 1 layer are surrounded by red frames F1, F1, and one second image region A2 where the number of stacking layers N2 is 2 layers is surrounded by a blue frame F2. Furthermore, the number of cargo 2 items M1, M2 and the number of stacking layers N1, N2 are superimposed as text information in each image region A1, A2.

[0120] As a result, the user of the image processing device 20 (for example, a customs inspector) can easily visually grasp the number M and stacking layers N of multiple cargo items 2 visible in the X-ray image, along with the X-ray image of the item being inspected, by looking at the displayed output image. Therefore, the user does not need to visually count the number M of cargo items 2 by observing the X-ray image as in the conventional method, thus reducing the burden on the user.

[0121] [8. Summary] The X-ray inspection apparatus 1 according to this embodiment has been described in detail above. According to this embodiment, a distribution of brightness values ​​of X-ray images taken of multiple cargo items 2 to be inspected is generated, and characteristic quantities of the cargo items to be inspected are extracted from the distribution of brightness values. Then, the number M of cargo items 2 to be inspected is calculated based on the characteristic quantities of the cargo items to be inspected and the reference characteristic quantities that have been registered in advance in the cargo DB 50 (storage unit).

[0122] This makes it possible to automatically and accurately identify the number M of cargo items 2 to be inspected. Therefore, as in the past, it is no longer necessary for users such as customs inspectors to visually count the number of cargo items 2 by observing X-ray images. Thus, even if the size of the container 3 containing cargo items 2 increases and the number of cargo items 2 that need to be counted visually increases, the burden on users can be greatly reduced.

[0123] Furthermore, according to this embodiment, when the object to be inspected consists of multiple cargo items 2 stacked in one or more layers, it is preferable to extract the characteristic quantities of the object to be inspected for each peak P1 and P2 of the histogram representing the distribution of brightness values ​​in the X-ray image of the object to be inspected. This makes it possible to accurately identify the number of cargo items Mn for each stacking layer Nn of cargo items 2.

[0124] Furthermore, according to this embodiment, it is preferable that the feature quantities to be inspected include the average brightness values ​​L1 and L2 of image regions A1 and A2 corresponding to each peak P1 and P2 of the histogram, and the areas S1 and S2 of the image regions. The average brightness values ​​L1 and L2 are the feature quantities to be inspected, and the average brightness value L is the reference feature quantity. ref Based on this, it is preferable to calculate the number of layers N1 and N2 of the stacked cargo 2. Furthermore, the calculated number of layers, the area S1 and S2 which are feature quantities of the object to be inspected, and the area S which is a reference feature quantity ref Based on this, it is preferable to calculate the number of cargo items M1 and M2 for each of the N1 and N2 layers. This makes it possible to more accurately determine the total number of cargo items M stacked in multiple layers.

[0125] Furthermore, according to this embodiment, it is preferable to generate an output image in which at least one of the information regarding the calculated number of cargo items M and the number of layers N in which the cargo items 2 are stacked is superimposed on the X-ray image of the object to be inspected. This allows the X-ray inspection device 1 to easily grasp the number of cargo items M and the number of layers N of the cargo items 2 to be inspected by looking at the displayed output image.

[0126] Furthermore, according to this embodiment, it is preferable that the cargo 2 is the same type of material packed in a container (for example, grain packed in a flexible container bag). This results in similar X-ray images of each individual cargo 2. Therefore, when calculating the number M of cargo 2 from X-ray images of multiple cargo 2 being inspected, it becomes possible to calculate the number more accurately.

[0127] While embodiments of this disclosure have been described above with reference to the attached drawings, it goes without saying that this disclosure is not limited to such embodiments. It will be obvious to those skilled in the art that various modifications or alterations can be conceived within the scope of the claims, and these will naturally also fall within the technical scope of this disclosure.

[0128] The series of processes performed by each device, such as the image processing apparatus 20 according to this embodiment, may be implemented using software, hardware, or a combination of software and hardware. The program constituting the software may be pre-stored, for example, in a non-transitory storage medium provided inside or outside each device. The program may then be read from a non-transitory storage medium (e.g., ROM) to a temporary storage medium (e.g., RAM) and executed by a processor such as a CPU.

[0129] It is possible to create programs to implement each of the above-mentioned devices and install them on the computers of each device. The processor executes the programs stored in memory, thereby carrying out the processing of each of the above-mentioned functions. At this time, the program may be divided and executed by multiple processors, or it may be executed by a single processor. Alternatively, each of the above-mentioned devices may be implemented through cloud computing, which uses multiple computers interconnected by a communication network.

[0130] The program may be provided to and installed on each device's computer via a communication network from an external device. Alternatively, the program may be stored on a non-transitory computer-readable medium and provided to and installed on each device's computer via that medium.

[0131] Furthermore, according to this embodiment, a program for executing the processing of each function of each of the above-mentioned devices can be provided. In addition, a non-temporary recording medium that can be read by a computer and on which the program is stored can also be provided. The non-temporary recording medium may be a disk-type recording medium such as an optical disk, magnetic disk, or magneto-optical disk, or it may be a semiconductor memory such as a flash memory or USB memory. [Explanation of symbols]

[0132] 1. X-ray inspection device 2 cargo 3 containers 5 X-ray 10. X-ray imaging device (imaging unit) 20 Image Processing Devices 30 Database Creation Department 32 Pre-processing section 34 Image Analysis Department 36 Feature Extraction Unit 38 Counting Section 40 Output image generation unit 50. Cargo Database (Storage Unit) Peaks in the P1, P2, and Pn histograms P1', P2', Pn' are peaks in the normal distribution. A1, A2, An image area

Claims

1. An imaging unit that images multiple cargo items to be inspected using X-rays and outputs X-ray images of the items to be inspected, An image analysis unit obtains the distribution of brightness values ​​of the X-ray image of the object to be inspected by analyzing the X-ray image of the object to be inspected, A feature extraction unit extracts the feature quantities of the object to be inspected based on the distribution of the brightness values, A storage unit that stores reference feature quantities, which are feature quantities obtained from the distribution of brightness values ​​of a reference X-ray image of a single cargo, A counting unit that calculates the number of the goods to be inspected based on the characteristic quantities of the items to be inspected and the reference characteristic quantities, Equipped with, The items to be inspected are those in which the aforementioned multiple cargoes are stacked in one or more layers. The X-ray image of the object to be inspected is a grayscale image and includes at least one image region with varying degrees of density depending on the number of layers in which the cargo is stacked. The image analysis unit generates a histogram representing the distribution of the brightness values ​​by analyzing the X-ray image of the object to be inspected. The histogram has at least one peak corresponding to the image region, where the intensity differs according to the number of steps. The feature extraction unit is an X-ray inspection apparatus that extracts the feature quantities of the object to be inspected for each peak in the histogram.

2. The X-ray inspection apparatus according to claim 1, wherein the feature quantities include the average brightness value of the image region corresponding to each peak of the histogram and the area of ​​the image region.

3. The aforementioned counting unit is, Based on the average brightness value, which is a feature of the object to be inspected, and the average brightness value, which is a reference feature, the number of layers of the stacked cargo is calculated. The X-ray inspection apparatus according to claim 2, which calculates the number of items of the cargo to be inspected based on the calculated number of stages, the area which is a characteristic quantity of the object to be inspected, and the area which is a standard characteristic quantity.

4. The X-ray inspection apparatus according to claim 1 or 2, further comprising an output image generation unit that generates an output image by superimposing on an X-ray image of the object to be inspected at least one of the following: information relating to the number of items of the cargo to be inspected calculated by the counting unit, and information relating to the number of layers in which the cargo to be inspected is stacked.

5. The X-ray inspection apparatus according to claim 1 or 2, wherein the cargo is the same type of material packaged in a container.

Images