CN104306009A - Radiographic image capturing apparatus - Google Patents

Abstract

The invention relates to a radiographic image capturing apparatus. To realize accurate detection for body motion of a subject when the subject can move during an imaging operation period such as long-length imaging, an image information acquisition unit acquires imaging situation during the imaging operation period and imaging information of the imaged subject; a local motion vector calculation unit calculates local motion vectors in an overlapping area between two adjacent radiographic images based on the imaging information; and body motion index value calculation unit calculates a body motion index value by using the local motion vectors based on the imaging information. Besides, a body motion determination unit determines whether the body motion exists or not by using the body motion index value based on the imaging information.

Description

Radiation image photography equipment

The divisional application that the application's to be the applying date be JIUYUE in 2011 8 days, application number are 201110264400.X, denomination of invention is the patent application of " health motion detection device and method, radiation image photography equipment and method ".

Technical field

The present invention relates to from the multiple radiation images obtained by carrying out repeatedly image photography operation to same subject, detect the health movement of this subject during photographing health motion detection device and method.

And, the present invention relates to by taking multiple radiation image and linking these radiation images to obtain radiation image photography equipment and the method for single long sized image (long-length image).

Background technology

Traditionally, in the fields such as medical treatment, propose and actually employed polytype radiation detector (so-called " flat-panel detector (Flat Panel Detectors) ", be expressed as hereinafter " FPD "), these radiation detectors record radiation image about subject by receiving transmitted through the lonizing radiation of subject.In these FPD, such as, there is the FPD of the quasiconductor using the amorphous selenium such as producing electric charge when being exposed to lonizing radiation.Propose the FPD of FPD and the TFT reading system of so-called optical access system as such FPD.

In addition, photographing relative to simple radioscopic image, the long size photography of the image of the long size area for taking such as whole spinal column or whole leg can being carried out, and have also already been the long size photography utilizing FPD.When carrying out the photography of long size, the region that FPD can be utilized to carry out taking may be narrower than the region of the subject expecting shooting.In this case, the position of FPD is moved along predetermined shifting axle, and the lonizing radiation that each shift position of FPD is all received transmitted through same subject are photographed to realize long size.During each irradiation lonizing radiation (when recording radiation image), read the image that is recorded from FPD to obtain the view data of the radiation image be recorded represented for each read operation at every turn.Afterwards, the view data of the radiation image obtained is carried out synthesize with by connected to each other for these images to provide the view data of the part of the length representing subject.

But, in the long size photography using FPD, interval of taking pictures be 3 seconds to 5 seconds, therefore, may there is health and move in subject between taking pictures.If subject has health to move between taking pictures, then can not correctly link obtained radioscopic image, thus hinder Measurement accuracy.Therefore, again image must be taken.In some cases, when observing the image of synthesis, operator notices that during photographing, there occurs health moves first, and after shooting full image, again take image is inefficent.In addition, the shooting carried out after health occurring and moves is otiose, and unnecessarily increases the chance that subject is exposed to lonizing radiation.

The problem of health movement also may occur during energy subtraction (energy subtraction) photography and temporal subtraction (temporal subtraction) photography, during energy subtraction photography, based on the fact that the decay of the lonizing radiation transmitted through subject changes according to the material of this subject of formation, two radiation images by irradiating the two bundle lonizing radiation acquisitions with different-energy to subject are utilized to carry out energy subtraction, and during temporal subtraction, obtain the difference image of the difference represented between two radiation images of different time shooting.In addition, the problem of health movement also or may occur during sequence photography during chromatography (tomosynthesis) photography, during tomographic, along with mobile X-ray tube is to take image to subject X-ray irradiation from different perspectives, and to provide, the image addition so obtained is emphasized that the image of the fault plane (slice plane) expected is to observe affected part (affected part) in more detail, and during sequence photography, obtain multiple image by irradiating lonizing radiation to subject continuously.

In order to address this problem, the health proposed for detecting subject during photographing moves the technology stopping photographing or warn health movement.Such as, Japanese Patent Laid-Open No.2009-240656 proposes a kind of technology, and this technology relates to the health utilizing sensor to detect subject during photographing and moves and give a warning when detecting that health moves.In addition, a kind of technology calculating the health amount of movement of subject when considering the alignment error of the rigging error of FPD and radiation image photography equipment is proposed.Particularly, propose so a kind of technology: perform template matching (template matching) in the overlapping region between multiple radiation image to find local displacement amount, this health amount of movement as the amount of movement of health, and is compared to detect with threshold value and whether be there is health and move by the average of displacement calculating amount.It should be noted that technology disclosed in Japanese Patent Laid-Open No.2009-240656 moves as health relative to the moving in parallel of photographing region of FPD for detecting subject.In addition, Japanese Patent Laid-Open No.2009-240656 proposes and utilizes non-linear warp to correct the technology of health movement.

But depend on the photography state of shooting operation and captured subject, for each shooting operation, health moves all exists change.Such as, taking larger interval photography between the image of larger amt and the image of longer photography time or shooting can cause larger health to move.In addition, the health that the health occurred when subject is not fixed moves than occurring when subject is fixed moves greatly.And, if the body part be taken comprises the bascule of such as heart, then very naturally there is larger health and move.On the contrary, if the body part be taken is leg, then health moves and can be reduced to minimum due to the self-control of the patient as subject.If subject is such as the patient be sent to hospital due to emergency or the patient just having experienced operation, then the situation of patient very critical and very difficult maintenance patient health be still in precalculated position.Therefore, according to the situation of subject, likely can not during photographing, keep health static, and large health may occur in this case move.And the health amount of movement required for health mobility detect can change according to subject.

Therefore, if carry out detection and the warning of health movement as technology disclosed in Japanese Patent Laid-Open No.2009-240656 according to changeless mode, then the possibility of result disunity of health mobility detect.Particularly, during the photography of chest, heart is in motion.If consider the motion of heart and arrange large threshold value, then may be difficult to detect that health moves when taking different parts.And, in the diagnosis to whole leg, measure the size of leg roughly.On the other hand, in the diagnosis to whole spinal column, do not measure the size of only spinal column, and measure the bending of spinal column.Therefore, for whole spinal column, detect that health moves required health amount of movement less than whole leg.But, if consider the degree of accuracy of the measurement of whole spinal column and arrange little threshold value, then during the long size photography of whole leg, detect that the little health not affecting measurement moves.

In addition, health moves not only disclosed in Japanese Patent Laid-Open No.2009-240656, comprise translation, but also comprise three-dimensional mobile, such as subject relative to the photographing region of FPD distortion and tilt forward or backward.In addition, health moves and comprises two-dimensional movement, zooming in or out of the rotation of such as subject in the plane parallel with photographing region and the subject image that causes in the translation forward or in backward directions due to subject.In technology disclosed in Japanese Patent Laid-Open No.2009-240656, the displacement only calculated on the direction parallel with photographing region is moved by as health, and therefore three-dimensional movement and two-dimensional movement translation are similar to.When in this manner with translation come approximate three-dimensional move with two-dimensional movement, even if subject exists be different from the larger movement of translation, testing result may only indicate little health move or do not have health to move.

On the other hand, if do not detect that health moves (if that is, health moves little), then apply non-linear warp and realize health shift calibrating.But, as mentioned above, though due to exist subject be different from translation large mobile time testing result little health also may be indicated to move, also carry out health shift calibrating even if therefore exist when comprising three-dimensional mobile and two-dimensional movement large mobile.If carry out health shift calibrating when there is subject large mobile, then thus image is produced large distortion by exaggerated correction, and this will hinder Accurate Diagnosis.

In addition, in technology above-mentioned disclosed in Japanese Patent Laid-Open No.2009-240656, warn when health occurring and moving, and operator takes image again when being warned.But, only can not just know that health moves according to warning and how to occur, and similar health moves and again may occur during again taking operation.If health again occurs during again taking operation move, be then necessary again to take image again, the amount of radiation causing inefficient photography there is no increasing subject necessarily and be subject to.

Summary of the invention

In view of said circumstances, the present invention is devoted to the accurate detection realizing the health movement to subject when subject is likely moved during the shooting operation of such as long size photography.

Health is there is and moves in the present invention during being also devoted to prevent occurring again to take operation after health moves in a front shooting operation during long size photography.

First aspect according to health motion detection device of the present invention comprises: image acquiring unit, and it carries out repeatedly shooting operation for same subject and the multiple radiation images taken for passing, and described radiation image is overlapping at least partly each other; Photographic information acquisition unit, it represents the photographic information of photography situation during shooting operation and captured subject for obtaining; And health moves desired value acquisition unit, it moves desired value for obtaining health based on described photographic information, and described health moves desired value and represents that the health of described subject during the described radiation image of shooting moves.

The expression of " overlapping at least partly " not only refers to the radiation image situation that part is overlapping each other, but also refers to the situation that radiation image is completely overlapping.

In the first aspect of health motion detection device according to the present invention, described health moves desired value acquisition unit and can comprise: local motion-vector calculating part, it is for calculating at least one local motion-vector, and this at least one local motion-vector represents the local displacement in the overlapping region of described subject between described radiation image; And health moves desired value calculating part, it moves desired value for calculating described health based on described local motion-vector.

In the case, described health moves the desired value that desired value calculating part can calculate the translation amount of described subject and moves desired value as described health.

Described health moves at least one that can also calculate in the desired value of the desired value of the three-dimensional amount of movement of described subject and the two-dimensional movement amount of described subject of desired value calculating part and moves desired value as described health.

Can also comprise health according to the first aspect of health motion detection device of the present invention and move judging part, described health moves judging part and moves for moving desired value to judge whether based on described photographic information and described health the health that there is described subject.

In the first aspect of health motion detection device according to the present invention, whether the interval photography between the photography time during described photographic information can comprise shooting operation, radiation image, described subject are fixed during shooting operation, at least one in the body part be taken of described subject and the symptom of described subject.

Second aspect according to health motion detection device of the present invention comprises: image acquiring unit, and it carries out repeatedly shooting operation for same subject and the multiple radiation images taken for passing, and described radiation image is overlapping at least partly each other; Photographic information acquisition unit, it represents the photographic information of photography situation during shooting operation and captured subject for obtaining; Health moves desired value acquisition unit, for obtaining, it represents that the health of the health movement of described subject during the described radiation image of shooting moves desired value; And health moves judging part, it moves for moving desired value to judge whether based on described photographic information and described health the health that there is described subject.

In the second aspect of health motion detection device according to the present invention, described health moves desired value acquisition unit and can comprise: local motion-vector calculating part, it is for calculating at least one local motion-vector, and at least one local motion-vector described represents the local displacement in the overlapping region of described subject between described radiation image; And health moves desired value calculating part, it moves desired value for calculating described health based on described local motion-vector.

In the case, described health moves the desired value that desired value calculating part can calculate the translation amount of described subject and moves desired value as described health.

Described health moves at least one that can also calculate in the desired value of the desired value of the three-dimensional amount of movement of described subject and the two-dimensional movement amount of described subject of desired value calculating part and moves desired value as described health.

In the second aspect of health motion detection device according to the present invention, whether the interval photography between the photography time during described photographic information can comprise shooting operation, radiation image, described subject are fixed during shooting operation, at least one in the body part be taken of described subject and the symptom of described subject.

A kind of radiation image photography equipment according to the first aspect of radiation image photography equipment of the present invention, this radiation image photography equipment is used for by the position of moving radiographic thread detector and irradiates lonizing radiation transmitted through described subject to obtain multiple radiation images overlapping at least partly each other when described position is moved to described radiation detector at every turn, this equipment comprises: photography portion, it is for moving the position of described radiation detector along predetermined shifting axle, and irradiate the lonizing radiation transmitted through described subject when position is moved to described radiation detector at every turn, radiation image acquisition unit, its for by each to be moved in position and lonizing radiation are illuminated time from described radiation detector read output signal obtain multiple radiation images of described subject, and the described health motion detection device of first aspect or second aspect.

First aspect according to health movement detection method of the present invention comprises the following steps: pass and carry out repeatedly shooting operation for same subject and the multiple radiation images taken, described radiation image is overlapping at least partly each other; Photography situation during acquisition expression shooting operation and the photographic information of captured subject; And moving desired value based on described photographic information acquisition health, described health moves desired value and represents that the health of described subject during the described radiation image of shooting moves.

Second aspect according to health movement detection method of the present invention comprises the following steps: pass and carry out repeatedly shooting operation for same subject and the multiple radiation images taken, described radiation image is overlapping at least partly each other; Photography situation during acquisition expression shooting operation and the photographic information of captured subject; Obtain and represent that the health of the health movement of described subject during the described radiation image of shooting moves desired value; And move desired value to judge whether based on described photographic information and described health the health that there is described subject and move.

It should be noted that can be provided by the form making computer perform the program of described method according to the first aspect of health movement detection method of the present invention and second aspect.

According to the first aspect of health motion detection device of the present invention and health movement detection method, obtain the photographic information representing photography situation during shooting operation and captured subject, and obtain based on described photographic information and represent that the health of the health movement of subject during the described radiation image of shooting moves desired value.Therefore, even if photography situation and captured subject change, also can obtain health exactly according to photography situation and captured subject and move desired value, thus achieve the accurate detection to health movement.

In addition, according to the second aspect of health motion detection device of the present invention and health movement detection method, obtain the photographic information representing photography situation during shooting operation and captured subject, and determine whether there is health based on this photographic information and move.Therefore, even if photography situation and captured subject change, also can determine whether there is health exactly according to photography situation and captured subject and move.

In addition, by calculating the local motion-vector representing the displacement of subject in overlapping region for each regional area in the overlapping region of radiation image, and calculate health based on this local motion-vector and move desired value, can by moving desired value relative to the simple acquisition health that calculates.

When the desired value of the amount of the translation by subject be calculated as health move desired value, achieve the detection of the health movement that the translation by subject is caused.

At least one in the desired value of the desired value of the three-dimensional amount of movement by subject and the two-dimensional movement amount of subject is calculated as health when moving desired value, achieve the three dimensional body movement of the such as distortion and inclination to subject and to the such as in rotary moving of subject and the detection of two-dimensional movement that zooms in or out.

When the first aspect of health motion detection device according to the present invention and health movement detection method and second aspect being applied to the photography of long size, achieve the accurate detection of the health movement to the subject occurred between the radiation image obtained being photographed by long size.

The third aspect according to health motion detection device of the present invention comprises: image acquiring unit, and it carries out repeatedly shooting operation for same subject and the multiple radiation images taken for passing, and described radiation image is overlapping at least partly each other; And health moves desired value acquisition unit, it moves desired value for obtaining dissimilar health according to the type of the movement of described subject during the described radiation image of shooting.

In the third aspect of health motion detection device according to the present invention, described health moves desired value acquisition unit and can comprise: local motion-vector calculating part, it is for calculating at least one local motion-vector, and at least one local motion-vector described represents the local displacement in the overlapping region of described subject between described radiation image; And health moves desired value calculating part, it moves desired value for calculating dissimilar health based on described local motion-vector.

In the case, described health moves at least two that desired value calculating part can calculate in the desired value of the desired value of the translation amount of described subject, the desired value of the three-dimensional amount of movement of described subject and the two-dimensional movement amount of described subject and moves desired value as described health.

The third aspect according to health motion detection device of the present invention can also comprise post processing selection portion, and described post processing selection portion is used for moving desired value or the described dissimilar health main health moved in desired value based on described dissimilar health and moves desired value to select post processing for described radiation image.

In health motion detection device according to a third aspect of the invention we, described post-processing detection device can move based on described dissimilar health the mobile type that desired value or described main health move desired value to determine the described subject being detected health movement, and selects described post processing according to the described result determined.

In the third aspect of health motion detection device according to the present invention, whether described post processing selection portion can be selected to described radiation image application health shift calibrating.

In the third aspect of health motion detection device according to the present invention, described post processing selection portion can select the type of the health shift calibrating that will be applied to described radiation image.

In the third aspect of health motion detection device according to the present invention, described post processing selection portion can select the health that will show to move the type of desired value.

A kind of radiation image photography equipment according to the second aspect of radiation image photography equipment of the present invention, this radiation image photography equipment is used for by the position of moving radiographic thread detector and irradiates lonizing radiation transmitted through subject to obtain multiple radiation images overlapping at least partly each other when described position is moved to described radiation detector at every turn, this equipment comprises: image photography portion, it is for moving the position of described radiation detector and irradiate the lonizing radiation transmitted through described subject when described position is moved to described radiation detector at every turn along predetermined shifting axle, radiation image acquisition unit, it is for by each multiple radiation images obtaining described subject when described position is moved and lonizing radiation are illuminated from described radiation detector read output signal, and health motion detection device according to a third aspect of the invention we.

The third aspect according to health movement detection method of the present invention comprises the following steps: pass and carry out repeatedly shooting operation for same subject and the multiple radiation images taken, described radiation image is overlapping at least partly each other; And obtain dissimilar health according to the type of movement of described subject during the described radiation image of shooting and move desired value.

The 3rd aspect according to health movement detection method of the present invention can be provided by the form making computer perform the program of described method.

According to the third aspect of health motion detection device of the present invention and health motion detection device method, obtain dissimilar health according to the type of the health movement of subject during the described radiation image of shooting and move desired value.Therefore, can move according to the health of subject and obtain suitable health and move desired value, realize health mobility detect accurately thus.In addition, move desired value based on the health detected, achieve the suitable selection to the post processing that will perform for radiation image in the stage afterwards.

In addition, by calculating the local motion-vector representing the displacement of subject in overlapping region for each regional area in the overlapping region of radiation image, and calculate health based on this local motion-vector and move desired value, can by moving desired value relative to the simple acquisition health that calculates.

At least two in the desired value of the desired value of the translation amount by subject, the desired value of the three-dimensional amount of movement of subject and the two-dimensional movement amount of subject are calculated as health when moving desired value, achieve the health of the translation by subject is moved, the three dimensional body of the such as distortion and inclination of subject moves and the detection of at least two in subject such as in rotary moving and the two-dimensional movement that zooms in or out.

When the third aspect of health motion detection device according to the present invention and health movement detection method being applied to the photography of long size, achieve the accurate detection of the health movement to the subject occurred between the radiation image obtained by long size photography.

A kind of radiation image photography equipment according to the third aspect of radiation image photography equipment of the present invention, this radiation image photography equipment is used for by the position of moving radiographic thread detector and irradiates lonizing radiation transmitted through subject to obtain multiple radiation images overlapping at least partly each other when position is moved to described radiation detector at every turn, this equipment comprises: photography portion, it is for moving the position of described radiation detector and irradiate the lonizing radiation transmitted through described subject to described radiation detector after position is moved at every turn along predetermined shifting axle, radiation image acquisition unit, its for by each to be moved in position and lonizing radiation are illuminated time to obtain multiple radiation images of described subject from described radiation detector read output signal, health mobility detect portion, it moves for detecting the health of described subject during the described radiation image of shooting, and information generation unit, it is for generating the rephotography auxiliary information of again taking for auxiliary described radiation image when detecting that health moves.

The third aspect according to radiation image photography equipment of the present invention can also comprise display part, and described display part is for showing described rephotography auxiliary information.

The third aspect according to radiation image photography equipment of the present invention can also comprise rephotography control part, and described rephotography control part is again taken described in controlling based on described rephotography auxiliary information.

In the third aspect of radiation image photography equipment according to the present invention, described rephotography control part can carry out frame distribution, takes described radiation image with the position of the generation health movement by avoiding described subject.

In the third aspect of radiation image photography equipment according to the present invention, described rephotography control part can increase the tightness of the fixture for fixing described subject.

In the third aspect of radiation image photography equipment according to the present invention, described health mobility detect portion can comprise: local motion-vector calculating part, it is for calculating at least one local motion-vector, and at least one local motion-vector described represents the local displacement in the overlapping region of described subject between described radiation image; Health moves desired value calculating part, and it moves desired value for calculating health based on described local motion-vector; And health moves judging part, there is health move for moving desired value to judge whether based on described health in it.

In the case, described health moves the desired value that desired value calculating part can calculate the amount of the translation of described subject and moves desired value as described health.

And in the case, described health moves at least one that can also calculate in the desired value of the desired value of the three-dimensional amount of movement of described subject and the two-dimensional movement amount of described subject of desired value calculating part and moves desired value as described health.

In the third aspect of radiation image photography equipment according to the present invention, described health mobility detect portion can be the sensor of the movement for detecting described subject.

Radiation image photography method according to the present invention is for irradiating lonizing radiation transmitted through subject to obtain the radiation image photography method of multiple radiation images overlapping at least partly each other to described radiation detector after position is moved by the position of moving radiographic thread detector at every turn, the method comprises the following steps: the position of moving described radiation detector along predetermined shifting axle, and after position is moved, irradiates the lonizing radiation transmitted through described subject to described radiation detector at every turn; Obtain multiple radiation images of described subject from described radiation detector read output signal during by moving in each position and irradiating lonizing radiation; Detect the health of described subject during the described radiation image of shooting to move; And the rephotography auxiliary information of again taking generated when detecting that health moves for auxiliary described radiation image.

Can be provided by the form making computer perform the program of described method according to radiation image photography method of the present invention.

According to the third aspect of radiation image photography equipment of the present invention and radiation image photography method, if detect that health moves, then generate rephotography auxiliary information.Therefore, operator can carry out rephotography operation according to rephotography auxiliary information, makes that health does not occur and moves.This makes it possible to effectively carry out rephotography operation and the exposure reducing subject.

Particularly, when showing rephotography auxiliary information, operator can promptly check rephotography auxiliary information, thus efficiently can carry out rephotography operation.

When carrying out rephotography operation based on rephotography auxiliary information, can rephotography operation be carried out, making not occur health and moving and not troublesome operations person.

In addition, by calculating the local motion-vector representing the displacement of subject in overlapping region for each regional area in the overlapping region of radiation image, calculate health based on this local motion-vector and move desired value, and move desired value based on this health to determine whether there is health and move, can by detecting health move relative to simple calculating.

When the desired value of the translation amount by subject be calculated as health move desired value, achieve the detection to the health movement that the translation by subject causes.

At least one in the desired value of the desired value of the three-dimensional amount of movement by subject and the two-dimensional movement amount of subject is calculated as health when moving desired value, and achieve the three dimensional body movement of the such as distortion and inclination to subject and subject is such as in rotary moving and the detection of two-dimensional movement that zooms in or out.

Accompanying drawing explanation

The schematic diagram of the structure of the radiation image photography equipment of Fig. 1 be exemplary application health motion detection device according to first embodiment of the invention;

Fig. 2 is the figure of the calculating for explaining local motion-vector;

Fig. 3 is the figure for explaining the template matching utilizing multiresolution to change;

Fig. 4 is the figure of the another kind of method for explaining template matching;

Fig. 5 is the figure of another method for explaining template matching;

Fig. 6 is the figure illustrating histogrammic example;

Fig. 7 is for explaining that the health of three-dimensional movement moves the figure of the calculating of desired value;

Fig. 8 is the figure for explaining health shift calibrating;

Fig. 9 is the flow chart illustrating the process carried out in the first embodiment;

Figure 10 is the figure of the example of display in illustration first embodiment on the display screen of image displaying part;

Figure 11 is the figure of another example of display in illustration first embodiment on the display screen of image displaying part;

The schematic diagram of the structure of the radiation image photography equipment of Figure 12 be exemplary application health motion detection device second embodiment of the invention;

Figure 13 is the flow chart illustrating the process carried out in this second embodiment;

Figure 14 has been exemplary application according to the schematic diagram of the structure of the radiation image photography equipment of the health motion detection device of the 3rd embodiment of the present invention;

Figure 15 is the flow chart illustrating the process carried out in the third embodiment;

Figure 16 is the figure of the example of display in illustration the 3rd embodiment on the display screen of image displaying part;

Figure 17 is the figure of another example of display in illustration the 3rd embodiment on the display screen of image displaying part;

Figure 18 has been exemplary application according to the schematic diagram of the structure of the radiation image photography equipment of the health motion detection device of the 4th embodiment of the present invention;

Figure 19 is the flow chart being illustrated in the process carried out in the 4th embodiment;

Figure 20 is the figure of the state illustrating the multiple composograph of display;

Figure 21 is the figure of the state illustrating the multiple composograph of display;

Figure 22 illustrates the schematic diagram according to the structure of the radiation image photography equipment of the 5th embodiment of the present invention;

Figure 23 is the flow chart being illustrated in the process carried out in the 5th embodiment;

Figure 24 is the figure of the example of the display illustrated when there is health and moving on the display screen of image displaying part;

Figure 25 is the figure of another example of the display illustrated when there is health and moving on the display screen of image displaying part;

Figure 26 is the figure of another example of the display illustrated when there is health and moving on the display screen of image displaying part;

Figure 27 is the figure of the Still another example of the display illustrated when there is health and moving on the display screen of image displaying part;

Figure 28 illustrates the schematic diagram according to the structure of the radiation image photography equipment of the 6th embodiment of the present invention;

Figure 29 is the flow chart being illustrated in the process carried out in the 6th embodiment; And

Figure 30 is the flow chart being illustrated in the process carried out in the 7th embodiment.

Detailed description of the invention

Below, embodiments of the present invention are described in detail with reference to accompanying drawing.The schematic diagram of the structure of the radiation image photography equipment of Fig. 1 be exemplary application health motion detection device according to first embodiment of the invention.As shown in Figure 1, multiple adjacent area N1 of subject N, N2 ... and link the radiation image so obtained and provide the long size radiation image of the major part representing subject N to carry out long size photography can in turn be taken by utilizing single radioactive source 100 and single FPD 110 according to the radiation image photography equipment 150 of the first embodiment.

The radiation image photography equipment 150 that it should be noted that according to the present embodiment not only can be used for the photography of long size, and may be used for general only to the photography of the specific part of the such as chest or leg of subject N.In the case, energy subtraction photography, temporal subtraction photography etc. can be carried out.But, in the following description, only relate to the photography of long size and describe the present invention in detail.

Particularly, radiation image photography equipment 150 comprises: radioactive source 100, and it is for launching lonizing radiation 104 by light exit window 111 to the exposure range limited by collimator 112; FPD 110, it has for receiving lonizing radiation 104 transmitted through subject N to detect the photographing region (lonizing radiation detection faces) 102 of lonizing radiation 104; Detector moving part 20, it is for moving FPD 110 along subject N; And radioactive source location division 25, its for located irradiation source 100 to make the position of light exit window 111 and towards being in the expectation state.In FIG, Cr represents the axis of the lonizing radiation 104 in the exposure range limited by collimator 112.

FPD 110 detects the lonizing radiation 104 transmitted through subject N and the lonizing radiation 104 detected is converted to the signal of telecommunication to export the view data representing the radiation image of subject N.FPD 110 can be the Direct-type FPD lonizing radiation detected directly being converted to electric charge, or first the lonizing radiation detected is converted to light and then light is converted to the indirect-type FPD of electric charge.Direct-type FPD is formed by the photoconduction electrolemma of such as amorphous selenium, capacitor and the TFT (thin film transistor (TFT)) etc. that is used as switching device.Such as, when the lonizing radiation of illuminated such as X-ray, produce electron-hole pair (e-h to) at photoconduction electrolemma.Electron-hole pair is stored in the capacitor, and the electric charge stored in the capacitor reads as the signal of telecommunication via TFT.

On the other hand, indirect-type FPD is made up of fluorescent material scintillator layers, photodiode, capacitor and TFT etc. are formed.When the lonizing radiation of illuminated such as " CsI:Tl ", scintillator layers luminescence (fluorescence).The fluorescence launched of scintillator layers through photodiode opto-electronic conversion and store in the capacitor, and store electric charge in the capacitor and read as the signal of telecommunication via TFT.

Detector moving part 20 comprises: two pillars 21, and they are upper upright to keep between which by FPD 110 in vertical direction (the arrow Y-direction accompanying drawing) from floor surface 5F; And travel mechanism 22, it is for mobile FPD 110 in vertical direction (longitudinal direction).Can use and utilize conventional known linear slide mechanism etc. support FPD 110 and use the drive source of such as motor to move the travel mechanism of FDP 110 as travel mechanism 22.

When performing the shooting operation obtaining the radiation image that will synthesize, subject N is located by the moving direction along FPD 110.That is, during photographing, subject N is erectly positioned on floor surface.

Radioactive source location division 25 keeps radioactive source 100 across the photographing region 102 (that is, towards FPD 110) of subject N towards FPD 110, and movable radiation source 100.Radioactive source location division 25 comprises: pillar 26, and it extends in vertical direction from ceiling 5E; Ceiling substrate 27, it moves pillar 26 along roof 5E in arrow Z-direction in the accompanying drawings; And turntable 28, it engages the side of arrow Y in the accompanying drawings can move up and can rotate around the axle vertical with paper with support column 26.Radioactive source 100 is arranged on turntable 28.Therefore, radioactive source 100 above can move at above-below direction (the arrow Y-direction in accompanying drawing) and left and right directions (the arrow Z-direction in accompanying drawing), and can rotate around parallel with the X-axis in figure and through the approximate centre of radioactive source 100 axle.Radioactive source location division 25 can also be formed by the drive source of known linear slide mechanism, rotating mechanism and the such as motor of routine.

Radiation image photography equipment 150 also comprises long size photography control part 50, and this long size photography control part 50 is for controlling the operation of detector moving part 20 and radioactive source location division 25.Long size photography control part 50 controls the operation of detector moving part 20 so that FPD 110 is sequentially moved to position Q1, Q2 ..., to carry out radiation image photography operation on the direction of subject N.Control in combination therewith, the operation that long size photography control part 50 controls radioactive source location division 25 is led by the photographing region 102 navigating to the FPD 110 of above-mentioned position successively radioactive source 100 to be orientated as the lonizing radiation 104 making to send from radioactive source 100.When driving radioactive source 100 in this case, in independent shooting operation, in turn take the adjacent region N1 of subject N, N2 ..., to provide multiple view data of the local radiation line image representing the image being used to form whole subject N.

Radiation image photography equipment 150 also comprises: health mobility detect portion 30, and it moves for detecting the health of subject N during shooting operation; Health shift calibrating portion 40, it is for moving the displacement correcting subject N based on health; Images uniting portion 42, it represents the long size radiation image of whole subject N with generation for the synthesis of the multiple view data obtained by above-mentioned radiation image photography operation; And warning portion 44.The long size radiation image generated by Images uniting portion 42 is presented on the image displaying part 60 that formed by such as CRT display device, liquid crystal indicator etc.

Health mobility detect portion 30 comprises image acquiring unit 31, photographic information acquisition unit 32, local motion-vector calculating part 34, health moves desired value calculating part 36 and health moves judging part 38.It should be noted that local motion-vector calculating part 34 and health move desired value calculating part 36 and define health and move desired value acquisition unit.

Image acquiring unit 31 is formed by the various interfaces for obtaining radiation image from FPD 110.It should be noted that image acquiring unit 31 is network interfaces when providing health mobility detect portion 30 discretely with radiation image photography equipment 150 and health mobility detect portion 30 is connected to radiation image photography equipment 150 via network.

Photographic information acquisition unit 32 obtains photographic information from the control station 70 of the integrated operation controlling radiation image photography equipment 150.Photographic information comprises about the photography information of situation and the information about captured subject.Taking the interval photography between photography time that whole radiation image spends, every two adjacent radiation images when can be included in carry out the photography of long size about the information of photography situation, whether fixing subject N etc. when photographing.Information about captured subject comprise the body part (such as chest, leg, whole spinal column, whole leg etc.) be taken of subject N and subject N symptom (such as situation is serious or slight, subject be before surgery or Post operation etc.).Photographic information acquisition unit 32 obtains at least one in these information as photographic information.

Should note, photographic information can be inputted via control station 70 by operator, or can by the control station 70 metric photography time, detect subject N and whether be fixed, identify that the body captured in obtained radiation image is graded and automatically calculated by control station 70.Because photographic information changes according to photography menu, the photography menu that control station 70 can be selected according to operator determines photographic information automatically.

Local motion-vector calculating part 34 calculates the local motion-vector in the overlapping region between every two adjacent radiation images.Fig. 2 is the figure of the calculating for explaining local motion-vector.As shown in Figure 2, (namely local motion-vector calculating part 34 mates the local amount of movement that occurs in the picture because the health of subject N moves in overlapping region K1 and K2 calculating two adjacent radiation image S1 and S2 by applying template, local motion-vector), in template matching, use the specific image part of one of them image as template (have and be present in the template T of the control point in the overlapping region K1 of radiation image S1 as benchmark) to find the image section (in the overlapping region K2 of radiation image S2) of the correspondence in another image.

Particularly, correlation between each object images I in the motion-vector calculating part 34 calculation template T and predetermined search ranges R of local, each object images I has the size same with template T-phase and is sequentially searched in the overlapping region K2 of radiation image S2.This operation provides the correlation distribution with the size identical with hunting zone R.Then, the pixel displacement amount (amount of movement) of the reference position (finding the position of maximum related value when not having health to move) of the position distance template T finding maximum related value in correlation distribution is calculated as local motion-vector V0.

It should be noted, in the present embodiment, be provided with multiple control point.Therefore, in overlapping region K1, multiple local motion-vector is calculated.Control point can be whole location of pixels of the image in overlapping region K1, or can be by the location of pixels that intended pixel compartment is dredged in overlapping region K1.Alternatively, control point can be the cross point at the edge in overlapping region K1, or representational location of pixels, such as has the location of pixels of large variance.Representational location of pixels can automatically detect, or manually can be arranged on the overlapping region K1 of shown radiation image S1 by operator.

Not only based on finding that the position of maximum related value calculates local motion-vector V0, but also can calculate local motion-vector V0 based on the position of centre of gravity of the correlation in correlation distribution.Calculate weighted average location with the initial point (such as the reference position of above-mentioned template) in the R of hunting zone for benchmark by the correlation based on the location of pixels in the R of hunting zone, the position of centre of gravity of correlation can be found.Now, correlation can be only used to be equal to, or greater than the location of pixels of predetermined threshold to calculate position of centre of gravity.This position of centre of gravity is called as " high relevant position of centre of gravity ".

In addition, as shown in Figure 3, when calculating local motion-vector V0, can to the image applications multiresolution conversion in overlapping region K1 and K2 with the overlapping region image producing different resolution, and the local motion-vector V0 between the overlapping region image that can in turn calculate each resolution according to the order from lowest resolution to highest resolution.It should be noted that Fig. 3 illustrates and carry out twice conversion of resolution to produce the state with the multi-resolution image dropping to 1/4 resolution.In addition, in figure 3, the overlapping region image K1-1 before multiresolution conversion and K2-1 labelling, the overlapping region image K1-2 of next resolution and K2-2 labelling, and the overlapping region image of next resolution uses K1-3 and K2-3 labelling respectively again.

Each conversion of resolution that multiresolution is changed creates the overlapping region image that resolution reduces by half.Therefore, assuming that calculate 16 local motion-vector V0 when application twice conversion of resolution is to realize being down to the multiple conversion of resolution of 1/4 resolution, first, overlapping region image K1-3 and K2-3 of lowest resolution is used to calculate a local motion-vector V0.It should be noted that when for explaining Fig. 3 of calculating each local motion-vector V0 a part of, for convenience's sake, showing the overlapping region image with different resolution identical with overlapping region K1-1, K2-1 dimensionally.

Subsequently, the second high-resolution overlapping region image K1-2 and K2-2 is used to calculate 4 local movement resolution V0.Now, by using the local motion-vector V0 calculated with overlapping region image K1-3 and K2-3, effective calculating of 4 local motion-vector V0 can be realized.Such as, as shown in Figure 3, when obliquely pointing to top-right motion-vector at the local motion-vector V0 calculated for overlapping region image K1-3 and K2-3, this local motion-vector V0 is used as the initial value of overlapping region image K1-2 and K2-2 of next resolution, to carry out template matching in the region (representing with hacures) only around the upper right quarter region of hunting zone R, as shown in Figure 4.This can reduce the amount of calculation for calculating correlation, realizes effective calculating of local motion-vector V0 thus.

Although in the above description hunting zone R is arranged on for carrying out in the overlapping region K2 of template matching, as shown in Figure 5, also hunting zone R can be arranged to exceed overlapping region K2.By increasing the size of hunting zone R in this way, the calculating more accurately to local motion-vector V0 can be realized.

In addition, although in the above description template T is arranged in the overlapping region K1 of radiation image S1 to calculate local motion-vector V0, but, except template T being arranged in overlapping region K1, template T can also be arranged in the overlapping region K2 of radiation image S2, calculate local motion-vector V0 to utilize overlapping region K2 as benchmark.

In addition, although use correlation to calculate local motion-vector V0 in the above description, any index of such as residual error or the mean square error etc. of the similarity between each object images I in beyond correlation, instruction template T and hunting zone R can be used.

As mentioned above, local motion-vector calculating part 34 can use in multiple method any one calculate local motion-vector V0; But in the present embodiment, local motion-vector calculating part 34 calculates local motion-vector V0 based on the photographic information that photographic information acquisition unit 32 obtains.Particularly, the method for calculating local motion-vector V0 changes according to photographic information.Such as, if about the information such as packets of information have taken great amount of images containing instruction, photography time is long, interval photography is long and subject is not fixed of photography situation, if or about captured subject packets of information containing the body part captured by instruction be chest (situation beyond the photography of long size) or whole spinal column (situation of long size photography), the critical and information that the health of patient can not be kept during photographing static etc. as the situation of the patient of subject, then may there is large health and move in supposition.Therefore, if moved assuming that may there is large health, then local motion-vector calculating part 34 can arrange large hunting zone R and/or drop to the overlapping region image of more low resolution to carry out template matching to the conversion of overlapping region application multiresolution to generate resolution.This makes it possible to calculate local motion-vector V0 more accurately.It should be noted that can use threshold value to determine photography time and whether grow and interval photography whether long.

On the contrary, move if the information supposition that can comprise according to photographic information may exist little health, then little hunting zone R can be set and/or reduce the number of times of multiresolution conversion or multiresolution conversion can not be applied.Except determining whether there is except large health moves, local motion-vector calculating part 34 can determine the amplitude of health movement step by step.In the case, according to the amplitude of health movement, the size/of hunting zone R can be changed step by step or change the number of times of conversion of resolution step by step.

In addition, if photographic information comprises the captured information comprising the body part of a lot of two-dimensional structure (such as chest, whole spinal column etc.), then can calculate each local motion-vector V0 based on the position of the maximum related value found during template matching.On the contrary, if photographic information comprises the captured information comprising the body part of a lot of one-dimentional structure (such as leg, whole leg etc.), then correlation trends towards along one-dimensional square to departing from.Therefore, each local motion-vector V0 can be calculated based on position of centre of gravity or high relevant position of centre of gravity.

Health moves desired value calculating part 36 and uses the local motion-vector V0 calculated by local motion-vector calculating part 34 to move desired value to calculate health.First, health moves the health that desired value calculating part 36 calculates translation and moves desired value.Translation refers to the Linear-moving parallel with the photographing region 102 of FPD 110 of subject N.Therefore, health moves desired value calculating part 36 and generates stereogram for the local motion-vector V0 that calculated by local motion-vector calculating part 34, wherein, and the length travel that three axles correspond to, lateral displacement and frequency.Length travel and lateral displacement correspond respectively to the vertical and horizontal of the overlapping region of radiation image.Use the coordinate system shown in Fig. 1, by represent with X-axis laterally and Y-axis represents that bivector on longitudinal X-Y coordinate is to represent each local motion-vector V0.Therefore, the length travel of each local motion-vector V0 and lateral displacement represent by the amplitude in the Y-direction of local motion-vector V0 and X-direction.

Fig. 6 is the figure illustrating histogrammic example.As shown in Figure 6, this rectangular histogram shows the frequency of the local motion-vector V0 of the value depending on length travel and lateral displacement.It should be noted that to generate rectangular histogram, only can use the local motion-vector V0 with following condition: have the basis being used as to calculate local motion-vector V0, to equal or higher than the correlation of predetermined threshold.Alternatively, can the variance yields of the pixel value for calculating each local motion-vector V0 in calculation template T, and only can use the variance yields utilizing and have and be less than predetermined threshold and the local motion-vector V0 that calculates of the template T comprising multiple limit.

Then, health moves desired value calculating part 36 and moves desired value based on histogram calculation health.Such as, by determining the local motion-vector V0 in rectangular histogram with maximum frequency, and the distance between the position (maximum frequency position) of local motion-vector V0 in compute histograms and reference position (i.e. histogrammic initial point) moves desired value as health, can calculate health and move desired value.Alternatively, by calculating the position of centre of gravity (frequency position of centre of gravity) with the local motion-vector V0 of the frequency being equal to, or greater than predetermined threshold, and then the distance calculated between frequency position of centre of gravity and reference position moves desired value as health, can calculate health and move desired value.Alternatively, can calculate the mean place that meansigma methods that the meansigma methods of the length travel of local motion-vector V0 and horizontal prosposition move calculates local motion-vector V0, and the distance that can calculate between this mean place and reference position moves desired value as health.Alternatively, the centre position that the intermediate value of length travel of local motion-vector V0 and the intermediate value of lateral displacement calculate local motion-vector V0 can be calculated, and the distance that can calculate between this centre position and reference position moves desired value as health.

In addition, health moves the health that desired value calculating part 36 calculates the three-dimensional movement of subject N and moves desired value.Three-dimensional mobile refer to subject N distortion and subject N relative to photographing region 102 inclination in the longitudinal direction of FPD 110.It is the histogram calculation using the local motion-vector V0 calculated as described above that the health of three-dimensional movement moves desired value.Fig. 7 is for explaining that the health of three-dimensional movement moves the figure of the calculating of desired value.It should be noted that when health move only comprise translation, histogrammic distribution roughly concentrates on a position.On the contrary, when health move comprise three-dimensional mobile, histogrammic distribution expansion.Therefore, health moves the standard deviation of the distribution in desired value calculating part 26 compute histograms or variance and moves desired value as the health of three-dimensional movement.Standard deviation or variance larger, three-dimensional mobile larger.Any one can be used in above-mentioned maximum frequency position, frequency position of centre of gravity, mean place and centre position as the center calculating standard deviation or variance.

In addition, health moves the health that desired value calculating part 36 calculates the two-dimensional movement of subject N and moves desired value.Two-dimensional movement refers to the rotation of subject N in the plane that the photographing region 102 with FPD 110 is parallel, the translation of subject N on the fore-and-aft direction relative to photographing region 102 and the translation of subject N on the direction parallel with photographing region 102.It should be noted that to move to look it is zooming in or out of radiation image at the health relative to the translation on the fore-and-aft direction of photographing region.Therefore, in the following description, the health on fore-and-aft direction is moved the health being described as zoom in/out to move.In addition, although local motion-vector V0 can be used as mentioned above to calculate the movement of translation, in this is to the description of the calculating of two-dimensional movement, suppose that two-dimensional movement comprises rotation, zoom in/out and translation whole.

Health moves desired value calculating part 36 such as based on " Zukei-shori-kogaku (graphic processing engineering) " (Fujio Yamaguchi, published by Nikkan Kogyo Shimbun Ltd., 68-82 page, 1981) health that equation disclosed in calculates the two-dimensional movement of subject N moves desired value.Particularly, the equation local motion-vector V0 calculated as described above being applied to two dimensional affine conversion moves desired value from each element of multiple local motion-vector V0 calculating translation, rotation and zoom in/out as health to use method of least square.Suppose two dimensional affine conversion in transverse shifting amount be tx and the amount of vertically moving time ty, translation is represented (in equation (1) and equation subsequently by equation (1) below, (x, y) be two-dimensional coordinate position before health moves, and (x*, y*) is health move after two-dimensional coordinate position):

$\left(\begin{array}{ccc}x& y& 1\end{array}\right)\left(\begin{array}{ccc}1& 0& 0\\ 0& 1& 0\\ \mathrm{tx}& \mathrm{ty}& 1\end{array}\right)=\left(\begin{array}{ccc}{\mathrm{xt}}^{*}& {\mathrm{yt}}^{*}& 1\end{array}\right)---\left(1\right)$

Suppose:

I＝(x y 1)

$\mathrm{Ht}=\left(\begin{array}{ccc}1& 0& 0\\ 0& 1& 0\\ \mathrm{tx}& \mathrm{ty}& 1\end{array}\right)$

It ^*＝(xt ^* yt ^* 1)，

Then equation (1) becomes:

IHt＝It ^* (1’)

In addition, suppose that the anglec of rotation in two dimensional affine conversion is θ, in rotary moving by equation (2) expression below:

$\left(\begin{array}{ccc}x& y& 1\end{array}\right)\left(\begin{array}{ccc}\cos \mathrm{\θ}& \sin \mathrm{\θ}& 0\\ -\sin \mathrm{\θ}& \cos \mathrm{\θ}& 0\\ x_{0\mathrm{\θ}}(1-\cos \mathrm{\θ})+y_{0\mathrm{\θ}}\sin \mathrm{\θ}& y_{0\mathrm{\θ}}(1-\cos \mathrm{\θ})-x_{0\mathrm{\θ}}\sin \mathrm{\θ}& 1\end{array}\right)=\left(\begin{array}{ccc}{x_{\mathrm{\θ}}}^{*}& {y_{\mathrm{\θ}}}^{*}& 1\end{array}\right)---\left(2\right)$

Wherein x _{0 θ}, y _{0 θ}it is center of rotation.

Suppose:

I＝(x y 1)

$H_{\mathrm{\θ}}=\left(\begin{array}{ccc}\cos \mathrm{\θ}& \sin \mathrm{\θ}& 0\\ -\sin \mathrm{\θ}& \cos \mathrm{\θ}& 0\\ x_{0\mathrm{\θ}}(1-\cos \mathrm{\θ})+y_{0\mathrm{\θ}}\sin \mathrm{\θ}& y_{0\mathrm{\θ}}(1-\cos \mathrm{\θ})-x_{0\mathrm{\θ}}\sin \mathrm{\θ}& 1\end{array}\right)$

I _θ ^*＝(x _θ ^* y _θ ^* 1)，

Then equation (2) becomes:

IH _θ＝I _θ ^* (2’)

In addition, suppose that the zoom in/out parameter of zoom in/out is a (X-direction) and d (Y-direction), the zoom in/out in two dimensional affine conversion is represented by equation (3) below:

$\left(\begin{array}{ccc}x& y& 1\end{array}\right)\left(\begin{array}{ccc}a& 0& 0\\ 0& d& 0\\ x_{0\mathrm{ad}}(1-a)& y_{0\mathrm{ad}}(1-d)& 1\end{array}\right)=\left(\begin{array}{ccc}{x_{\mathrm{ad}}}^{*}& {y_{\mathrm{ad}}}^{*}& 1\end{array}\right)---\left(3\right)$

Wherein x _0ad, y _0adit is center of rotation.

Suppose:

I＝(x y 1)

$H_{\mathrm{ad}}=\left(\begin{array}{ccc}a& 0& 0\\ 0& d& 0\\ x_{0\mathrm{ad}}(1-a)& y_{0\mathrm{ad}}(1-d)& 1\end{array}\right)$

I _ad ^*＝(x _ad ^* y _ad ^* 1)，

Then equation (3) becomes:

I _ad ^*＝(x _ad ^*y _ad ^* 1)，

IH _ad＝I _ad ^* (3’)

Merge equation (1 ') to (3 '), obtain equation (4) below:

IH＝I ^*

H＝Ht H _θ H _ad (4)

Wherein, H is the matrix comprising 9 elements.Local motion-vector V0 can be expressed as (x*-x, y*-y).Therefore, by finding nine elements to above equation (4) application local motion-vector V0 to utilize method of least square, translation amount, the anglec of rotation and Scalable parameter can be calculated.The health that the translation amount so calculated, the anglec of rotation and Scalable parameter are used as two-dimensional movement moves desired value.Should note, in the operation subsequently of present embodiment, the health of the translation gone out based on histogram calculation as described above moves the health that desired value is used as translation and moves desired value, and the health that only anglec of rotation and zoom in/out parameter are used as two-dimensional movement moves desired value.

Although any one health calculating translation and three-dimensional movement that health moves that desired value calculating part 36 can use in above-mentioned multiple method as described above moves desired value, the health in present embodiment moves desired value calculating part 36 and calculates health based on the photographic information that photographic information acquisition unit 32 obtains and move desired value.Particularly, the method moving desired value for calculating health changes according to photographic information.Such as, health for translation moves desired value, if captured body part is leg (situation beyond long size photography) or whole leg (situation of long size photography), the health then do not existed independent of subject N moves and the structure of movement, the gas in such as heart or intestinal.Therefore, if the photographic information body part comprised captured by instruction is the information of leg, then utilize mean place and centre position to move desired value to calculate health, utilize local motion-vector to carry out parameter value thus as much as possible.Health for three-dimensional movement moves desired value, utilizes and moves desired value as the mean place at the center for calculating standard deviation or variance and centre position to calculate health.

On the other hand, if captured body part is chest (situation beyond long size photography) or whole spinal column (situation of long size photography), the health then existed independent of subject N moves and the structure of movement, the gas in such as heart or intestinal.Therefore, if the photographic information body part comprised captured by instruction is the information of chest, then utilize maximum frequency position or frequency position of centre of gravity to move desired value to the health calculating translation, thus not by moving influence ground, local parameter value.Health for three-dimensional movement moves desired value, utilizes and moves desired value as the maximum frequency position at the center for calculating standard deviation or variance or frequency position of centre of gravity to calculate health.

Although health moves, desired value calculating part 36 calculates translation in this example, whole healths that are three-dimensional mobile and two-dimensional movement move desired value, health moves that desired value calculating part 36 can calculate translation, three-dimensional health that is mobile and two-dimensional movement moves in desired value at least one.Preferably, the health of translation moves desired value and is included.

Health moves judging part 38 and health is moved health that desired value calculating part 36 calculates and move desired value and threshold value compares, if health moves desired value and is equal to, or greater than threshold value, then health moves judging part 38 and judges to there is health and move and export judged result.Should note, if calculate translation, health that is three-dimensional mobile and two-dimensional movement moves in desired value two or more, then health moves judging part 38 and each health is moved desired value and compare to judge whether that there is health moves with corresponding threshold value.In the case, if any one movement in desired value of health is equal to, or greater than threshold value, then judge that there is health moves.Alternatively, if health moves in desired value, any two are equal to, or greater than threshold value, or all health moves desired value and is all equal to, or greater than threshold value, then can judge that there is health moves.In addition, can judge whether that there is health moves for each in translation, three-D displacement and two-dimension displacement.In addition, for two-dimensional movement, can prepare for the threshold value rotated and the threshold value for zoom in/out, and can judge whether that there is health moves for each in rotation and zoom in/out.

In the present embodiment, health moves photographic information that judging part 38 calculates based on photographic information acquisition unit 32 and moves desired value to health and threshold value compares, thus judges whether that there is health moves.Particularly, the size of threshold value changes according to photographic information.Such as, if captured body part is leg (situation beyond long size photography) or whole leg (situation of long size photography), the health then do not existed independent of subject N moves and the structure of movement, the gas in such as heart or intestinal.On the contrary, if captured body part is chest (situation beyond long size photography) or whole spinal column (situation of long size photography), the health then existed independent of subject N moves and the structure of movement, the gas in such as heart or intestinal.Therefore, if the photographic information body part comprised captured by instruction is the information of chest, then the threshold value that the threshold value used when to use than captured body part be leg is large judges.Like this, judgement about whether there is health movement can not be made by local moving influence.

In addition, if captured body part is the body part needing the health movement detected comparatively in a small amount, such as whole spinal column (situation of long size photography), then arrange less threshold value.On the contrary, if captured body part is the body part needing to detect relatively large health movement, such as whole spinal column (situation of long size photography), then arrange larger threshold value.

Below, the health shift calibrating performed by health shift calibrating portion 40 is described.If the judged result that health moves judging part 38 there is not health to move, then health shift calibrating portion 40 moves desired value based on the health detected by health mobility detect portion 30 and corrects radiation image to eliminate the anamorphose caused because the health of subject N moves.On the contrary, if judged result there is health to move, then do not carry out health shift calibrating, because if carry out health shift calibrating, then anamorphose is more serious.Below, the correction for eliminating the anamorphose between two adjacent radiation images is described.As shown in Figure 8 A, when detecting that the health of translation moves desired value, as shown in Figure 8 B, desired value is moved (namely based on health, local motion-vector V0), by relative to each other moving two radiation image S1 and S2, achieve health shift calibrating.When detecting that two or more health moves desired value, by relative to each other non-linearly warpage radiation image, make template aligned with each other with the object images being detected displacement, thus realize health shift calibrating, that is, targeted by the position of the local motion-vector V0 between two images.

When calculating translation, three-dimensional whole healths that are mobile and two-dimensional movement move desired value, 0 (namely if it is not that whole healths moves desired value, there is the health of at least two comprised in translation, three-dimensional movement and two-dimensional movement to move), then can realize health shift calibrating by carrying out non-linear warp.If translation, health that is three-dimensional mobile and two-dimensional movement moves in desired value only has one not to be 0, then can perform the health shift calibrating according to movement.Such as, if to move desired value non-vanishing for the health of only translation, then by relative to each other moving radiographic line image, health shift calibrating can be realized.Such as, if to move desired value non-vanishing for the health of only three-dimensional movement, then pass through non-linear warp, health shift calibrating can be realized.If only to move desired value non-vanishing for the health of two-dimensional movement, then by using health to move desired value (namely, rotation except equation (1) to the translation in (3) and the element of zoom in/out) two dimensional affine conversion, health shift calibrating can be realized.If the health of two-dimensional movement to move in desired value and only to rotate and health one of moving in desired value of zoom in/out is not 0, then by use this be not 0 the health two dimensional affine that moves desired value convert, health shift calibrating can be realized.If the health that there is two-dimensional movement moves, then non-linear warp can be carried out.In addition, can move according to three-dimensional and carry out health shift calibrating with the movement that the health maximum health moved in desired value moves desired value that has in two-dimensional movement.

Images uniting portion 42 synthesizes radiation image through health shift calibrating to produce composograph C1 by linking image.

If there is large health to move, then as will be described below, warning portion 44 warns.

The integrated operation of radiation image photography equipment 150 is controlled by control station 70.Therefore, about subject N, be imported into control station 70 for the information obtaining the photography situation of long size radiation image etc., and these information are imported into long size photography control part 50 with photography adjustment part (not shown) to arrange the radiation exposure scope etc. limited by collimator 112.Photography adjustment part carry out during each shooting radiation image frame divide be equipped with adjustment radioactive source 100 position, the situation of collimator 112, the position etc. of FPD 110, make the radiation image that will be synthesized all obtaining preliminary dimension in the operation each time in such as four radiation image photographies operations.Afterwards, according to the order inputted from control station 70, perform the operation of shooting four radiation images.

In order to determine the size of the radiation image captured by shooting operation, except limiting except radiation exposure scope with collimator 112 as described above, a part for the radiation image that each shooting operation obtains can be excised to adjust size and the width of image section.

Next, the process performed in the first embodiment will be described.Fig. 9 is the flow chart illustrating the process performed in the first embodiment.First, by mobile FPD 110 to take radiation image along mobile route in each position, long size photography (step ST1) is performed.Then, the photographic information acquisition unit 32 in health mobility detect portion 30 obtains photographic information (step ST2) from control station 70.Subsequently, local motion-vector calculating part 34 calculates local motion-vector V0 (step ST3) in the overlapping region between every two adjacent radiation images based on photographic information, and health moves desired value calculating part 36 calculates health based on photographic information and move desired value (step ST4).Then, health moves judging part 38 and moves desired value based on photographic information and health and judge whether that there is health moves (step ST5).

If judge there is not health and move, then health moves judging part 38 and judged result and health is moved together with desired value and output to health shift calibrating portion 40 and image displaying part 60 (step ST6).Health shift calibrating portion 40 moves desired value based on health and to move the health in radiation image and correct (step ST7), and Images uniting portion 42 synthesizes radiation image through health shift calibrating to produce composograph C1 (step ST8).Then, composograph C1 moves together with desired value with health and shows (step ST9) by image displaying part 60, and this process terminates.

Figure 10 is the figure of the example of display on the display screen of the image displaying part 60 of illustration first embodiment.As shown in Figure 10, the health that display screen 61 comprises the image display area 62 for showing composograph C1 and moves desired value for showing health moves desired value viewing area 63.When calculating moves desired value more than the health of a type, display on desired value viewing area 63 can be moved at health and move desired value with each health be associated in translation, three-dimensional movement and two-dimensional movement.Alternatively, the health that only can show translation moves desired value, or translation can be shown, maximum health that health that is three-dimensional mobile and two-dimensional movement moves in desired value moves desired value.In the case, owing to moving desired value use different units for translation, three-dimensional health that is mobile and two-dimensional movement, therefore can by these unit-normalization to determine that health moves the size of desired value.Health for two-dimensional movement moves desired value, only can show one that rotation and the health of zoom in/out move in desired value, can show whole health and move desired value or only can show maximum health and move desired value.Alternatively, as shown in figure 11, time near the overlapping region that cursor moves to the composograph C1 of display in image display area 62, health moves desired value and can show in pop-up window.

On the contrary, if judge that in step ST5 there is health moves, then health moves judging part 38 and judged result is outputted to warning portion 44 (step ST10).Warning portion 44 utilizes speech message (comprising voice warning) or buzzer sound (warning sound) to carry out warning (step ST11), and this process terminates.In the case, speech message can be the speech message that prompting operator takes image again, as " there is large health to move.Please again take image ".Alternatively, warning label or alert message can be shown on the display screen of image displaying part 60.Alternatively, sound and display two kinds of modes can be used to warn.Alternatively, judged result and health move desired value can be output to image displaying part 60, and only health move desired value can according to judge that the mode that there is not the situation of health movement identical is presented on image displaying part 60.

As mentioned above, in the present embodiment, the photographic information acquisition instruction health that subject is taking the health movement during radiation image based on the photography situation represented during shooting operation and captured subject moves desired value, and judges whether that there is health moves based on photographic information.Therefore, even if when photography situation and captured subject change, still can obtain health exactly move desired value and can judge whether that there is health moves exactly according to photography situation and captured subject, thus achieve to the accurate detection of health movement with to the accurate judgement that whether there is health movement.

Next, the second embodiment of the present invention is described.The schematic diagram of the structure of the radiation image photography equipment of Figure 12 be exemplary application health motion detection device second embodiment of the invention.It should be noted that parts identical with the first embodiment in the second embodiment represent with same reference numerals and are not described in detail.Be according to the difference between the radiation image photography equipment 150A of the second embodiment and the radiation image photography equipment of the first embodiment, in this second embodiment, only the health in health mobility detect portion 30 moves judging part 38 and judges whether that there is health moves based on the photographic information that photographic information acquisition unit 32 obtains.That is, in this second embodiment, local motion-vector calculating part 34 and health move desired value calculating part 36 and calculate local motion-vector V0 according to preordering method and health moves desired value and do not use photographic information.

Next, the process performed in the second embodiment of the present invention will be described in.Figure 13 is the flow chart illustrating the process performed in this second embodiment.First, by mobile FPD 110 to take radiation image along mobile route in each position, executive chairman's size photography (step ST21).Then, the photographic information acquisition unit 32 in health mobility detect portion 30 obtains photographic information (step ST22) from control station 70.Subsequently, local motion-vector calculating part 34 calculates local motion-vector V0 (step ST23) in the overlapping region between every two adjacent radiation images, and health moves desired value calculating part 36 calculates health and move desired value (step ST24).Then, health moves judging part 38 and moves desired value based on photographic information and health and judge whether that there is health moves (step ST25).

If judge there is not health and move, then health moves judging part 38 and judged result and health is moved together with desired value and output to health shift calibrating portion 40 and image displaying part 60 (step ST26).Health shift calibrating portion 40 moves desired value based on health and to move the health in radiation image and correct (step ST27), and Images uniting portion 42 synthesizes radiation image through health shift calibrating to generate composograph C1 (step ST28).Then, composograph C1 moves together with desired value with health and shows (step ST29) by image displaying part 60, and this process terminates.

On the contrary, if judge that in step ST25 there is health moves, then health moves judging part 38 and judged result is outputted to warning portion 44 (step ST30).Warning portion 44 speech message (comprising voice warning) or buzzer sound (warning sound) carry out warning (step ST31), and this process terminates.

Should note, in the first embodiment described before, although local motion-vector calculating part 34 calculates local motion-vector V0 based on photographic information and health moves desired value calculating part 36 calculates health based on photographic information and move desired value, local motion-vector calculating part 34 and health move in desired value calculating part 36 only has one can calculate based on photographic information.In addition, in the first embodiment described before, health move judging part 38 do not use photographic information just can judge whether there is health moves.In the case, the threshold value for judging can be fixed in advance.

Next, the 3rd embodiment of the present invention is described.Figure 14 has been exemplary application according to the schematic diagram of the structure of the radiation image photography equipment of the health motion detection device of the 3rd embodiment of the present invention.It should be noted that parts identical with the first embodiment in the 3rd embodiment represent with same reference numerals and are not described in detail.Be according to the difference between the radiation image photography equipment that radiation image photography equipment 150B and first enforcement of the 3rd embodiment occurs, in the third embodiment, post processing selection portion 39 is set in health mobility detect portion 30 and moves judging part 38 to replace health.

It should be noted that in the third embodiment, local motion-vector calculating part 34 is not limited to calculate local motion-vector V0 based on photographic information as described above.Can pre-determine local motion-vector calculating part 34 uses any method in said method to calculate local motion-vector V0.In the case, photographic information acquisition unit 32 is optional.

In addition, health moves that desired value calculating part 36 is not limited to calculate translation, whole healths that are three-dimensional mobile and two-dimensional movement move desired value in the third embodiment.At least two that health moves that desired value calculating part 36 can calculate translation, whole healths that are three-dimensional mobile and two-dimensional movement move in desired value.But preferably, the health of translation moves desired value and is included.

In addition, in the 3rd embodiment, health moves desired value calculating part 34 and is not limited to calculate health based on photographic information as described above and moves desired value.Health can be pre-determined move desired value calculating part 34 and use any method in said method to move desired value to calculate health.In the case, photographic information acquisition unit 32 is optional.

What post processing selection portion 39 calculated as described above based on the movement according to subject moves desired value to select the process that will perform in later phases more than the health of a type.For this reason, first post processing selection portion 39 moves desired value and threshold value compares to being moved each health that desired value calculating part 36 calculates by health, if health moves desired value be equal to, or greater than threshold value, then judges that there is health moves.In the third embodiment, calculate translation, health that is three-dimensional mobile and two-dimensional movement moves desired value, and desired value is moved to each health compare with corresponding threshold value, judge whether that there is health moves to move desired value for each health.For two-dimensional movement, prepare for the threshold value of each in rotation and zoom in/out to judge whether that there is health moves for each in rotation and zoom in/out.

In the third embodiment, post processing selection portion 39 compares judge whether that there is health moves by moving desired value and threshold value based on the photographic information calculated by photographic information acquisition unit 32 to each health.Particularly, the size of threshold value changes according to photographic information.Such as, if captured body part is leg (situation beyond long size photography) or whole leg (situation of long size photography), the health then do not existed independent of subject N moves and the structure of movement, the gas in such as heart or intestinal.On the contrary, if captured body part is chest (situation beyond long size photography) or whole spinal column (situation of long size photography), the health then existed independent of subject N moves and the structure of movement, the gas in such as heart or intestinal.Therefore, if the photographic information body part comprised captured by instruction is the information of chest, then the threshold value that the threshold value used when to use than captured body part be leg is large judges.Like this, can not judge whether that there is health moves by local moving influence.

It should be noted that in the third embodiment, by using threshold value fixing in advance, can when not based on judging whether when photographic information that there is health moves.In the case, photographic information acquisition unit 32 is optional.

In addition, post processing selection portion 39 is selected whether to perform health shift calibrating by health shift calibrating portion 40.Particularly, if judge that the health that there is three-dimensional movement moves, then the health that whether there is translation moves to move with the health of two-dimensional movement and does not carry out health shift calibrating.If judge that the health that there is two-dimensional movement moves, then the health that whether there is translation moves to move with the health of three-dimensional movement and does not carry out health shift calibrating.Health for translation moves, if judge that there is not health that is three-dimensional mobile and two-dimensional movement moves, then whether there is health and moves and all carry out health shift calibrating.That is, except the situation except there is the health movement of any one in three-dimensional movement and two-dimensional movement, health shift calibrating is carried out.

In addition, if select to carry out health shift calibrating by moving desired value based on the judged result and health that whether there is health movement, then post processing selection portion 39 selects the method for realizing health shift calibrating.It should be noted that in the present embodiment, desired value is moved to each health and compares with corresponding threshold value, if health moves desired value be less than threshold value, then judge that there is not health moves.Therefore, even if when judging that there is not health moves, health moves desired value and differs to be decided to be 0 and may to there is health and move.If the health that there is three-dimensional movement moves, then select non-linear warp.If the health that there is the rotation of two-dimensional movement moves, then select according to above equation (2) by rotating the method realizing health shift calibrating.If the health that there is the zoom in/out of two-dimensional movement moves, then select the method realizing health shift calibrating according to above equation (3) by zooming in or out.It should be noted that when the health that there is two-dimensional movement moves, can the method for choice for use non-linear warp.If there is three-dimensional health that is mobile and these two kinds of modes of two-dimensional movement to move, then the method for choice for use non-linear warp.When in addition to the above, select for the direction based on local motion-vector V0 by moving in parallel the health shift calibrating method realized for translation.

In addition, post processing selection portion 39 moves desired value based on moving desired value to select the health that will show on image displaying part 60 to the result and health that whether there is the judgement of health movement, and it will in following description.Particularly, if judge that the health that there is three-dimensional movement moves, even if if or the health that there is three-dimensional movement when judging that the health that there is not three-dimensional movement moves move, then the health that selection shows three-dimensional movement moves desired value (i.e. standard deviation or variance).If judge that the health that there is the rotation of two-dimensional movement moves, if even if or there is when the health judging not exist rotation moves the health rotated and move, then select the health of display rotation to move desired value (i.e. anglec of rotation θ).If judge that the health that there is the zoom in/out of two-dimensional movement moves, even if if or when judging that the health that there is not zoom in/out exists zoom in/out health when moving moves, then select the health of display zoom in/out to move desired value (i.e. parameter a and d).Health for translation moves, and selects the health of display translation to move desired value.

In the third embodiment, if post processing selection portion 39 has been selected to carry out health shift calibrating, then health shift calibrating portion 40 has performed to correct to radiation image by the health shift calibrating method that uses post processing selection portion 39 and select and has moved with the health eliminated due to subject N and the anamorphose that causes.When selecting to carry out health shift calibrating for translation, the health of translation moves desired value and is detected, as shown in Figure 8 A.Therefore, as shown in Figure 8 B, relative to each other move two radiation image S1 and S2 by moving desired value (that is, local motion-vector V0) based on health, achieve health shift calibrating.When selecting non-linear warp, by making radiation image relative to each other non-linear warp, achieve health shift calibrating.When selecting to carry out health shift calibrating for rotation or zoom in/out, carry out health shift calibrating according to above-mentioned equation (2) or (3).

Next, the process performed in the third embodiment will be described.Figure 15 is the flow chart illustrating the process performed in the third embodiment.First, by mobile FPD 110 to take radiation image along mobile route in each position, executive chairman's size photography (step ST41).Then, local motion-vector calculating part 34 calculates local motion-vector V0 (step ST42) in the overlapping region between every two adjacent radiation images, and health moves desired value calculating part 36 calculates health and move desired value (step ST43).

Subsequently, post processing selection portion 39 moves desired value based on health and judges whether that there is health moves (step ST44).Then, post processing selection portion 39 moves desired value to select the health that will show on image displaying part 60 based on the judged result and health that whether there is health movement and moves desired value (step ST45), and selects whether to carry out health shift calibrating (step ST46) according to the judged result that whether there is health movement.If make affirmative judgement in step ST46, then post processing selection portion 39 selects the method (step ST47) for realizing health shift calibrating, and respectively the health being used for realizing health shift calibrating moved desired value and output to health shift calibrating portion 40 about the information of selected health shift calibrating method, the selected health that will show is moved desired value and output to image displaying part 60 (output information, step ST48).

Then, the health shift calibrating method that health shift calibrating portion 40 uses post processing selection portion 39 to select moves the health in radiation image and corrects (step ST49), and Images uniting portion 42 synthesizes radiation image through health shift calibrating to produce composograph C1 (step ST50).Then, composograph C1 moves together with desired value with the health that will be shown by selection and shows (step ST51) by image displaying part 60, and this process terminates.

Figure 16 is the figure of the example of display on the display screen of image displaying part 60 in illustration the 3rd embodiment.As shown in figure 16, the health that display screen 61 comprises the image display area 62 for showing composograph C1 and moves desired value for showing health moves desired value viewing area 63.It should be noted that moving in desired value viewing area 63 at health the health only shown by selection will show moves desired value.Figure 16 shows the state that display translation, the three-dimensional mobile and rotation of two-dimensional movement and the health of zoom in/out move desired value.It should be noted that as shown in figure 17, time near the overlapping region that cursor moves to the composograph C1 of display in image display area 62, health moves desired value and can show in pop-up window.On the contrary, if make negative judgment in step ST46, then post processing selection portion 39 exports the selection result that health shift calibrating is not carried out in instruction to warning portion 44, and moves desired value (output information, step ST52) to the health that image displaying part 60 exports by selection will show.Warning portion 44 speech message (comprising voice warning) or buzzer sound (warning sound) carry out warning (step ST53).And image displaying part 60 shows is selected the health that will show to move desired value (step ST54), and this process terminates.In the case, speech message can be the speech message that prompting operator takes image again, as " there is large health to move.Please again take image ".Alternatively, warning label or alert message can be shown on the display screen of image displaying part 60.Alternatively, can warn by sound and display two kinds of modes.

As mentioned above, in the third embodiment, move according to the health of subject during radiation image photography operation, obtain and move desired value more than the health of a type.Therefore, the health according to subject moves, and can obtain translation, three-dimensional suitable health that is mobile and two-dimensional movement moves desired value, realize the accurate detection to health movement thus.In addition, move desired value based on the health detected, the stage to whether afterwards that can suitably realize carry out health shift calibrating selection, to the selection of the health shift calibrating method that will use and the selection to the health shift calibrating value that will show.

Next, the 4th embodiment of the present invention is described.Figure 18 has been exemplary application according to the schematic diagram of the structure of the radiation image photography equipment of the health motion detection device of the 4th embodiment of the present invention.It should be noted that parts identical with the 3rd embodiment in the 4th embodiment represent by same reference numerals and are not described in detail.Difference according to the radiation image photography equipment 150C of the 4th embodiment and the radiation image photography equipment of the 3rd embodiment is, in the 4th embodiment, post processing selection portion 39A determines translation, the health maximum health moved in desired value that is three-dimensional mobile and two-dimensional movement moves desired value and moves desired value as main health, and moves desired value to realize to the selection whether carrying out health shift calibrating, to the selection of the health shift calibrating method that will use and the selection of the health that will show being moved to desired value based on this main health.

That is, in the 4th embodiment, post processing selection portion 39A determines by the size that health moves the translation that desired value calculating part 36 calculates, three-dimensional health that is mobile and two-dimensional movement moves desired value.In the case, owing to moving desired value employ different units for translation, three-dimensional health that is mobile and two-dimensional movement, therefore can by these unit-normalization to determine that health moves the size of desired value.Then, maximum health is wherein moved desired value and be defined as main health and move desired value, and move desired value to select whether carry out health shift calibrating based on this main health.That is, if main health moves desired value is three-dimensional mobile or the health of two-dimensional movement moves desired value, then this health is moved desired value and threshold value compares to select whether to carry out health shift calibrating.If it is that the health of translation moves desired value that main health moves desired value, if then determine that there is not three-dimensional health that is mobile and two-dimensional movement moves, then health shift calibrating is carried out in selection.

In addition, if select to carry out health shift calibrating, then the method for realizing health shift calibrating is selected.That is, if main health moves desired value is that the health of three-dimensional movement moves desired value, then non-linear warp is selected.If it is that the health of the rotation of two-dimensional movement moves desired value that main health moves desired value, then select according to above-mentioned equation (2) by rotating the method realizing health shift calibrating.If it is that the health of the zoom in/out of two-dimensional movement moves desired value that main health moves desired value, then select the method being realized health shift calibrating according to above-mentioned equation (3) by zoom in/out.It should be noted that when the health that there is two-dimensional movement moves, can the method for choice for use non-linear warp.If it is that the health of translation moves desired value that main health moves desired value, then select to move in parallel by the direction based on local motion-vector V0 the method realized for the health shift calibrating of translation.

Below, the process performed in the 4th embodiment of the present invention will be described in.Figure 19 is the flow chart being illustrated in the process performed in the 4th embodiment.First, by mobile FPD 110 to take radiation image along mobile route in each position, executive chairman's size photography (step ST61).Subsequently, local motion-vector calculating part 34 calculates local motion-vector V0 (step ST62) in the overlapping region between every two adjacent radiation images, and health moves desired value calculating part 36 calculates health and move desired value (step ST63).

Subsequently, the size that post processing selection portion 39 pairs of healths move desired value compares to determine that main health moves desired value (step ST64) mutually.In addition, post processing selection portion 39 moves desired value to judge whether based on this main health to be existed health and moves (step ST65).Then, post processing selection portion 39 is selected main health to move desired value and is moved desired value (step ST66) as the health that will show on image displaying part 60, and selects whether to carry out health shift calibrating (step ST67) according to the result moving desired value for this main health and judge whether to exist health movement.If make affirmative judgement in step ST67, then post processing selection portion 39 moves desired value to select the method (step ST68) for realizing health shift calibrating based on this main health, and respectively, this main health is moved desired value and output to health shift calibrating portion 40 about the information of selected health shift calibrating method, the selected essential body that will show is moved desired value and output to image displaying part 60 (output information, step ST69).

Then, health shift calibrating portion 40 utilizes the health shift calibrating method selected by post processing selection portion 39 to move the health in radiation image and corrects (step ST70), and Images uniting portion 42 synthesizes radiation image through health shift calibrating to produce composograph C1 (step ST71).Then, composograph C1 moves together with desired value with the health that will be shown by selection and shows (step ST72) by image displaying part 60, and this process terminates.

On the contrary, if make negative judgment in step ST67, then post processing selection portion 39 exports the selection result that health shift calibrating is not carried out in instruction to warning portion 44, and moves desired value (output information, step ST73) to the main health that image displaying part 60 exports by selection will show.Warning portion 44 speech message (comprising voice warning) or buzzer sound (warning sound) carry out warning (step ST74).And image displaying part 60 shows is selected the health that will show to move desired value (step ST75), and this process terminates.

It should be noted that, in the 3rd embodiment described before, post processing selection portion 39 can select the health shift calibrating method more than a type.Such as, if exist comprise whole translation, health that is three-dimensional mobile and two-dimensional movement moves, then select to comprise move in parallel, non-linear warp, whole health shift calibrating methods of rotating and zooming in or out.It should be noted that for two-dimensional movement, sometimes can select non-linear warp.Therefore, health shift calibrating portion 40 can use the health shift calibrating method selected by post processing selection portion 39 whole may combine or combination in any to carry out health shift calibrating.Such as, if selecting to comprise moves in parallel, non-linear warp, the whole health shift calibrating methods rotating and zoom in or out, then there is the health shift calibrating method of this Four types, and by using a kind of health shift calibrating method of whole Four types (combination), use three kinds (four kinds of various combinations) in the health shift calibrating method of Four types, use two kinds (six kinds of various combinations) in the health shift calibrating method of Four types, or use the one (four kinds of various combinations or pattern) in the health shift calibrating method of Four types.That is, the various combination of 15 kinds of health shift calibrating methods or pattern can be used for realizing health shift calibrating altogether.

In the case, health shift calibrating portion 40 can use whole in 15 kinds of patterns of health shift calibrating method or wherein any amount kind pattern to realize health shift calibrating.Then, Images uniting portion 42 utilizes the whole radiation images crossed according to the mode correction of applied different health shift calibratings to generate multiple composograph.Then, image displaying part 60 shows multiple composograph.Figure 20 shows the state showing multiple composograph.It should be noted that Figure 20 shows the six kinds of patterns and the situation showing six composographs performing health shift calibrating.As shown in figure 20, the image display area 62 of display screen 61 comprise with tabular form display composograph reduce view.In the case, the composograph showing and choose can preferably be amplified.This makes operator can select one according to the corrected composograph of preference pattern of health shift calibrating method.In the case, as shown in figure 21, the composograph being different from the composograph corrected by the most preferred mode of health shift calibrating method can become ash (grey out).As shown in figure 21, shade indicates and turns grey state.

Should note, in above-mentioned 3rd embodiment and the 4th embodiment, although post processing selection portion 39 is selected whether to carry out health shift calibrating, the health shift calibrating method that selection will use also selects the health that will show on image displaying part 60 to move desired value, but post processing selection portion 39 can select any one or combination in any in these items.

Next, the 5th embodiment of the present invention is described.Figure 22 illustrates the schematic diagram according to the structure of the radiation image photography equipment of the 5th embodiment of the present invention.It should be noted that parts identical with the first embodiment in the 5th embodiment represent by same reference numerals and are not described in detail.Be according to the difference between the radiation image photography equipment 150D of the 5th embodiment and the radiation image photography equipment of the first embodiment, in the 5th embodiment, be provided with rephotography auxiliary information generating unit 46.

If health moves judging part 38 judge that there is health moves, then rephotography auxiliary information generating unit 46 generates and is used for auxiliary rephotography auxiliary information of again taking image.In order to realize the photography of long size, moving if there is large health during shooting operation, being then necessary again to take image.Therefore, move the judged result of judging part 38 based on health, the information that there is the adjacent radiation image of two of health movement between rephotography auxiliary information generating unit 46 generating identification and the information that will show on image displaying part 60 move as rephotography auxiliary information so that with the health of reduction takes again.The information that will show on image displaying part 60 can comprise the information of such as word, image (comprising still image and live image) and sound, is fixed by the body part of the generation health movement of subject N to point out operator.Should note, when judging that there is the information that will show when health moves can be such information, this message advises operator carries out single shooting operation by using the long size case of single large FPD or CR etc., instead of carries out the long size photography of taking multiple radiation image.Can writings and image be prepared in advance and be stored in storage part (not shown).

As mentioned above in the 5th embodiment, calculate health based on photographic information move desired value although health moves desired value calculating part 36, what can pre-determine that health moves that desired value calculating part 36 uses in said method anyly calculates health and moves desired value.In the case, photographic information acquisition unit 32 is optional.

Next, the process performed in the 5th embodiment will be described in.Figure 23 is the flow chart being illustrated in these process in the 5th embodiment.First, by mobile FPD 110 to take radiation image along mobile route in each position, executive chairman's size photography (step ST81).Subsequently, local motion-vector calculating part 34 calculates local motion-vector V0 (step ST82) in the overlapping region between every two adjacent radiation images, and health moves desired value calculating part 36 calculates health and move desired value (step ST83).Then, health moves judging part 38 and judges whether that there is health moves (step ST84).

If judge there is not health and move, then health moves judging part 38 and judged result and health is moved together with desired value and output to health shift calibrating portion 40 and image displaying part 60 (step ST85).Health shift calibrating portion 40 moves desired value based on this health and to move the health in radiation image and correct (step ST86), and Images uniting portion 42 synthesizes radiation image through health shift calibrating to generate composograph C1 (step ST87).Then, composograph C1 moves together with desired value with health and shows (step ST88) by image displaying part 60, and this process terminates.

Composograph C1 and health move desired value and are similarly presented on image displaying part 60 according to the state shown in Figure 10 or Figure 11 in the above-described first embodiment.

On the contrary, if judge that in step ST84 there is health moves, then health moves judging part 38 and judged result is outputted to warning portion 44 and rephotography auxiliary information generating unit 46 (step ST89).Rephotography auxiliary information generating unit 46 generates rephotography auxiliary information (step ST90), and this rephotography auxiliary information is outputted to image displaying part 60.Image displaying part 60 shows this rephotography auxiliary information (step ST91).In addition, warning portion 44 speech message (comprising voice warning) or buzzer sound (warning sound) carry out warning (step ST92), and this process terminates.In the case, speech message can be the speech message that prompting operator takes image again, such as " there is large health to move.Please again take image ".

Figure 24 is the figure of the example of the display illustrating rephotography auxiliary information.As shown in figure 24, the display screen 61 of image displaying part 60 shows image display area 64, this image display area 64 schematically shows the multiple radiation images (in this example, four radiation images) obtained by long size photography.As shown in figure 24, add instruction to four radiation image S1 to the S4 be presented on display screen 61 and the arrow of the position of health movement occurs as rephotography auxiliary information.The position of arrow instruction between radiation image S1 and radiation image S2 shown in Figure 24, and can see that between radiation image S1 and radiation image S2, health occurring moves.Symbol etc. can be used to replace arrow.In addition, as shown in figure 24, word 65 " health between the first image and the second image moves " is shown as rephotography auxiliary information.The operator observing this rephotography auxiliary information can notify subject N, the fixture of such as belt is used to fix the body part corresponding with the position between the first image and the second image of subject, and/or fix body part by placing mat, make between the first image and the second image, health not to occur during rephotography operation and move.

As shown in figure 25, time near the position that cursor moves to the arrow of increase, word can be shown in pop-up window.Alternatively, as shown in figure 26, schematic human body image 66 can be shown replace four radiation image S1 to S4 on image displaying part 64, and arrow can be arranged on the position of the human body image 66 that health movement occurs.

Except word or replace word, the securing guidance diagram of body part for pointing out there is health movement can be shown.Figure 27 shows the state of display guidance diagram.In figure 27, guidance diagram shows together with word.As shown in figure 27, image display area 64 is provided with the arrow of the position of instruction between radiation image S3 and radiation image S4.Position between radiation image S3 and radiation image S4 corresponds to the position of the knee of subject N.Therefore, the content of word 65 is " please fixing knee with belt ", and instructs Figure 67 to show the state using belt to fix knee, wherein, with belt 68 movable along arrow A direction in animated display guide image 67.

It should be noted that when rephotography auxiliary information suggestion operations person uses the long size case of single large FPD or CR etc. to carry out single shooting operation, the content of word can be " come in row single shooting operation ".

As mentioned above, in the 5th embodiment, if judge that there is health moves, then generate rephotography auxiliary information.Therefore, operator can carry out rephotography operation according to rephotography command information, makes that health does not occur and moves.This makes it possible to effectively carry out rephotography operation and the exposure reducing subject.Particularly, when showing rephotography auxiliary information, operator can promptly check rephotography auxiliary information, more effectively can carry out rephotography operation thus.

Then, the 6th embodiment of the present invention is described.Figure 28 has been exemplary application according to the schematic diagram of the structure of the radiation image photography equipment of the health motion detection device of the 6th embodiment of the present invention.It should be noted that parts identical with the 5th embodiment in the 6th embodiment represent by same reference numerals and are not described in detail.Be according to the difference between the radiation image photography equipment 150E of the 6th embodiment and the radiation image photography equipment of the 5th embodiment, in the 6th embodiment, be provided with rephotography control part 48.

Rephotography control part 48 orders above-mentioned Image Adjusting portion again to carry out frame distribution based on the rephotography auxiliary information generated by rephotography auxiliary information generating unit 46, to avoid the position that health movement occurs.It should be noted that under radiation image photography equipment 150E comprises the fixture of such as belt for fixing subject N and the tightness of this fixture is adjustable situation, the order that this fixture of can sening as an envoy to is tighter.Hereinafter, this instruction will be called " rephotography control ".

Below, the process performed in the 6th embodiment of the present invention will be described in.Figure 29 is the flow chart being illustrated in the process performed in the 6th embodiment.First, by mobile FPD 110 to take radiation image along mobile route in each position, executive chairman's size photography (step ST101).Subsequently, local motion-vector calculating part 34 calculates local motion-vector V0 (step ST102) in the overlapping region between every two adjacent radiation images, and health moves desired value calculating part 36 calculates health and move desired value (step ST103).Then, health moves judging part 38 and judges whether that there is health moves (step ST104).

If judge there is not health and move, then health moves judging part 38 and judged result and health is moved together with desired value and output to health shift calibrating portion 40 and image displaying part 60 (step ST105).Health shift calibrating portion 40 moves desired value based on health and to move the health in radiation image and correct (step ST106), and Images uniting portion 42 synthesizes radiation image through health shift calibrating to generate composograph C1 (step ST107).Then, composograph C1 moves together with desired value with health and shows (step ST108) by image displaying part 60, and this process terminates.

On the contrary, if judge that in step ST104 there is health moves, then health moves judging part 38 and judged result is outputted to warning portion 44 and rephotography auxiliary information generating unit 46 (step ST109).Rephotography auxiliary information generating unit 46 generates rephotography auxiliary information (step ST110), and this rephotography auxiliary information is outputted to image displaying part 60 and rephotography control part 48.Image displaying part 60 shows rephotography auxiliary information (step ST111).In addition, rephotography control part 48 carries out above-mentioned rephotography control (step ST112).On the other hand, warning portion 44 speech message (comprising voice warning) or buzzer sound (warning sound) carry out warning (step ST113), and this process terminates.

It should be noted that in the above-described 6th embodiment, although show rephotography auxiliary information on image displaying part 60, can only carry out rephotography control and not show rephotography auxiliary information.

At the first embodiment in the 6th embodiment, although calculate translation, three-dimensional health that is mobile and two-dimensional movement move desired value, but the health of known subject during particularly long size photography move mainly subject relative to FPD 110 horizontal and vertical on movement, that is, translation.By relative to each other moving radiographic line image, the health that can correct translation moves, and this does not make the image quality decrease of radiation image.Therefore, the health of known translation move be in do not affect measure and diagnosis scope in time, then the correction carried out is moved on the health of translation and any impact is not applied on the image interpretation of the composograph obtained by synthesis radiation image.If can know during the photography of long size health move be translation and this health move be in do not affect measure and diagnosis scope in, then because the image interpretation of the correction not synthetic images of radiation image applies any impact, thus without the need to again taking radiation image, as a result, subject can be prevented unnecessarily to be exposed to lonizing radiation.

For this reason, when calculating health and moving desired value, the health that only can calculate translation moves desired value.Below, this aspect is described as the 7th embodiment.Difference between 7th embodiment and above-mentioned embodiment is only to move by health the operation that desired value calculating part 36 carries out, and is therefore described in detail the structure of the 7th embodiment.

In the 7th embodiment, health moves the health that desired value calculating part 36 only calculates translation and moves desired value.Particularly, the rectangular histogram shown in Fig. 6 is that the local motion-vector V0 using partial vector calculating part 34 to calculate generates, and the health of translation moves desired value calculates based on this rectangular histogram.If captured body part is chest (situation beyond long size photography) or whole spinal column (situation of long size photography), the health then existed independent of subject N moves and the structure of movement, the gas in such as heart or intestinal.That is, the local included except movement relevant to the attitude of this subject N when subject N is considered to rigid body is moved.Health moves desired value calculating part 36 and uses the maximum frequency position in the rectangular histogram generated or frequency position of centre of gravity to move desired value to calculate health.It should be noted that and the weighted mean of compute histograms can move desired value as health, wherein, give larger weight to the local motion-vector with larger frequency.

Like this, can not move desired value by the health locally only calculating translation, this translation is the movement relevant to the attitude of this subject N when subject N is considered to rigid body moving influence.

The method that health for calculating translation moves desired value is not limited to use histogrammic method.Such as, desired value can be moved by as follows the minimum health calculating translation being reduced in the impact of local movement, the method comprises the following steps: assembled multiple local motion-vector V0 for the direction local motion-vector group identical with size is to be separated local motion-vector V0, and from the local motion-vector V0 be separated, remove the larger weight of the local motion-vector V0 that is less than predetermined threshold or the larger local motion-vector V0 of imparting.

Next, the process performed in the 7th embodiment will be described in.Figure 30 is the flow chart being illustrated in the process performed in the 7th embodiment.First, by mobile FPD 110 to take radiation image along mobile route in each position, executive chairman's size photography (step ST121).Then, local motion-vector calculating part 34 calculates local motion-vector V0 (step ST122) in the overlapping region between every two adjacent radiation images, and health moves the health that desired value calculating part 36 calculates translation moves desired value (step ST123).

Subsequently, post processing selection portion 39 moves desired value to judge whether based on health to be existed health and moves (step ST124).Then, post processing selection portion 39 is selected whether to carry out health shift calibrating (step ST125) according to the judged result that whether there is health movement.If make affirmative judgement in step ST125, then the information that the health of translation is moved desired value by post processing selection portion 39 outputs to health shift calibrating portion 40 respectively and arrives image displaying part 60 (information exports, step ST126).

Then, the health of the translation in 40 pairs, health shift calibrating portion radiation image moves and corrects (step ST127), and Images uniting portion 42 synthesizes radiation image through health shift calibrating to generate composograph C1 (step ST128).Then, composograph C1 moves together with desired value with the health of translation and shows (step ST129) by image displaying part 60, and this process terminates.

On the contrary, if make negative judgment in step ST125, then post processing selection portion 39 exports instruction and does not carry out the selection result of health shift calibrating to warning portion 44, and moves desired value (output information, step ST130) to the health that image displaying part 60 exports translation.Warning portion 44 speech message (comprising voice warning) or buzzer sound (warning sound) carry out warning (step ST131).And the health that image displaying part 60 shows translation moves desired value (step ST132), and this process terminates.

In the case, in the 7th embodiment, the accurate detection of the health movement of the translation to subject can be realized.In addition, desired value can be moved based on the health of the translation detected and suitably realize the selection whether afterwards stage to carries out health shift calibrating.

At above-mentioned first embodiment in the 4th embodiment and the 7th embodiment, desired value is moved by the health of the radiation image of long size photography shooting although obtain, but according to the mode similar with the 4th embodiment to above-mentioned 3rd embodiment, health mobility detect portion 30 can separately for moving desired value to calculate locally motion-vector V0 and health, and based on about to carry out same subject, the shooting operation of the health movement of subject may be there is (such as between taking pictures, energy subtraction is photographed, temporal subtraction is photographed, tomographic and continuously shooting) captured by the photographic information of radiation image judge whether that there is health moves.

In addition, at above-mentioned first embodiment in the 4th embodiment and the 7th embodiment, when mobile FPD 110 (, during photographing), health move desired value calculating and can based on each position in second position and position subsequently (namely to the judgement that whether there is health movement, shooting operation each time in second time shooting operation and later shooting operation) every two adjacent radiation images between overlapping region carry out, if judge that there is health moves, then can warn.And, in above-mentioned 5th embodiment and the 6th embodiment, if judge that there is health moves, then can carry out warning and can generating rephotography auxiliary information.

In the case, when operator response in warn etc. identify the health of subject N during the photography of long size move time, operator can stop shooting operation by the emergency stop switch of operation setting on control station 70.Even if when there is too large health and move to such an extent as to shooting image can not being synthesized, also prevent and cause subject N to be unnecessarily exposed to lonizing radiation owing to proceeding shooting operation.In addition, in above-mentioned 5th embodiment and the 6th embodiment, the health during rephotography auxiliary information helps operator to take measures to prevent again taking operation moves.

In the above-described embodiment, although judging to exist health moves during long size photography to warn, when warning, shooting operation can be automatically stopped.Which eliminate when operator response in warn wait and the health that identifies subject moves time operator immediate operation emergency stop switch needs, or this prevent the situation of carrying out next shooting operation before operation emergency stop switch, prevent subject be unnecessarily exposed to lonizing radiation and burden can not be brought to operator thus.

In addition, in above-mentioned 5th embodiment and the 6th embodiment, detect health move desired value although local motion-vector calculating part 34 and health move desired value calculating part 36, sensor can be set at detector moving part 20 place and move to detect health.Such as, sensor can be arranged on the position of placing with the sensor of distortion, the left and right foot that is arranged on subject of waving that handrail (not shown) sentences the health detecting subject to have the sensor that weight claims or the position sensor be arranged on subject N with what detect the change of body weight.

Describe equipment according to the embodiment of the present invention.The present invention can also realize with the form of program, and this program makes computer play with the function of above-mentioned photographic information acquisition unit 32, locally motion-vector calculating part 34, health moves desired value calculating part 36, health moves judging part 38, post processing selection portion 39, rephotography auxiliary information generating unit 46 and rephotography control part 48 are corresponding device to perform the operation shown in Fig. 9, Figure 13, Figure 15, Figure 19, Figure 23, Figure 29 or Figure 30.In addition, the present invention can also realize with the form comprising the computer readable recording medium storing program for performing of this program.

Embodiments of the invention are described below.

Embodiment 1

A kind of health motion detection device, this health motion detection device comprises:

Image acquiring unit, it carries out repeatedly shooting operation for same subject and the multiple radiation images taken for passing, and these radiation images are overlapping at least partly each other;

Local motion-vector calculating part, it is for calculating at least one the local motion-vector of the local displacement represented in the overlapping region of subject between described radiation image; And

Health moves desired value calculating part, and it moves desired value for calculating based on local motion-vector the health caused by the attitude of subject.

" health caused by the attitude of subject moves desired value " refers to the desired value of the health movement caused by the attitude of this subject when subject is considered to a rigid body, and do not comprise by subject body all after one's own heart, the internal such as lung, intestinal and the desired value of the movement caused.

Embodiment 2

As the health motion detection device that embodiment 1 limits, wherein, this health moves the desired value of translation amount that desired value calculating part calculates subject and moves desired value as health.

Embodiment 3

As the health motion detection device that embodiment 2 limits, wherein, this health moves at least one calculating in the desired value of three-dimensional amount of movement of subject and the desired value of the two-dimensional movement amount of subject of desired value calculating part and moves desired value as health.

Embodiment 4

As the health motion detection device that embodiment 3 limits, this health motion detection device also comprises post processing selection portion, and this post processing selection portion is for moving desired value to select the post processing for radiation image based on this health.

Embodiment 5

As the health motion detection device that embodiment 4 limits, wherein, whether this post processing selection portion is selected to radiation image application health shift calibrating.

Embodiment 6

As the health motion detection device that embodiment 4 or 5 limits, wherein, this post processing selection portion selects the type that will be applied to the health shift calibrating of radiation image.

Embodiment 7

The health motion detection device limited any one of embodiment 4 to 6, wherein, this post processing selection portion selects the health that will show to move the type of desired value.

Embodiment 8

The health motion detection device limited any one of embodiment 1 to 7, this health motion detection device also comprises photographic information acquisition unit, and this photographic information acquisition unit represents the photographic information of photography situation during shooting operation and captured subject for obtaining.

Embodiment 9

As the health motion detection device that embodiment 8 limits, wherein, this local motion-vector calculating part calculates at least one local motion-vector based on photographic information.

Embodiment 10

As the health motion detection device that embodiment 8 or 9 limits, wherein, this health moves the health that desired value calculating part calculates translation by least one local motion-vector of unifying based on this photographic information and moves desired value.

Embodiment 11

The health motion detection device limited any one of embodiment 8 to 10, this health motion detection device also comprises health and moves judging part, and this health moves judging part and moves for moving desired value to judge whether based on photographic information and health the health that there is subject.

Embodiment 12

The health motion detection device limited any one of embodiment 1 to 10, this health motion detection device also comprises health and moves judging part, and this health moves judging part and moves for moving desired value to judge whether the health that there is subject based on health.

Embodiment 13

A kind of radiation image photography equipment, this radiation image photography equipment is by the position of moving radiographic thread detector and each lonizing radiation irradiated transmitted through subject to radiation detector when position is moved obtain multiple radiation images overlapping at least partly each other, and this radiation image photography equipment comprises:

Photography portion, its for along predetermined shifting axle moving radiographic thread detector position and each irradiate the lonizing radiation transmitted through subject when position is moved to radiation detector;

Radiation image acquisition unit, it is for by each multiple radiation images obtaining subject when position is moved and lonizing radiation are illuminated from radiation detector read output signal; And

The health motion detection device limited any one of embodiment 1 to 12.

Embodiment 14

As the radiation image photography equipment that embodiment 13 limits, this radiation image photography equipment also comprises information generation unit, and this information generation unit is used for the rephotography auxiliary information of again taking generating auxiliary radiation line image when detecting that health moves.

Embodiment 15

As the radiation image photography equipment that embodiment 14 limits, this radiation image photography equipment also comprises display part, and this display part is for showing rephotography auxiliary information.

Embodiment 16

As the radiation image photography equipment that embodiment 14 or 15 limits, this radiation image photography equipment also comprises rephotography control part, and this rephotography control part is used for controlling again to take based on rephotography auxiliary information.

Embodiment 17

A kind of health movement detection method, this health movement detection method comprises the following steps:

Pass and carry out repeatedly shooting operation for same subject and the multiple radiation images taken, these radiation images are overlapping at least partly each other;

Calculate at least one the local motion-vector of the local displacement represented in the overlapping region of subject between radiation image; And

Calculate based on this local motion-vector the health caused by the attitude of subject and move desired value.

Embodiment 18

Make computer perform a program for health movement detection method, this health movement detection method comprises the following steps:

Obtain the multiple radiation images taken by carrying out repeatedly shooting operation to same subject, these radiation images are overlapping at least partly each other;

Calculate at least one the local motion-vector of the local displacement represented in the overlapping region of subject between radiation image; And

Calculate based on local motion-vector the health caused by the attitude of subject and move desired value.

Claims (9)

1. a radiation image photography equipment, this radiation imaging apparatus is used for the position by moving radiographic thread detector and irradiates the lonizing radiation transmitted through subject when described position is moved to described radiation detector at every turn, obtain multiple radiation images overlapping at least partly each other, this radiation image photography equipment comprises:

Photography portion, it is for moving the position of described radiation detector and irradiate the lonizing radiation transmitted through described subject when described position is moved to described radiation detector at every turn along predetermined shifting axle;

Radiation image acquisition unit, it is for by each multiple radiation images obtaining described subject when described position is moved and lonizing radiation are illuminated from described radiation detector read output signal;

Health mobility detect portion, it moves for detecting the health of described subject during the described radiation image of shooting; And

Information generation unit, it is for generating the rephotography auxiliary information of again taking for auxiliary described radiation image when detecting that health moves.

2. radiation image photography equipment according to claim 1, this radiation image photography equipment also comprises display part, and described display part is for showing described rephotography auxiliary information.

3. radiation image photography equipment according to claim 1, this radiation image photography equipment also comprises rephotography control part, and described rephotography control part is again taken described in controlling based on described rephotography auxiliary information.

4. radiation image photography equipment according to claim 3, wherein, described rephotography control part carries out frame distribution, takes described radiation image with the position of the generation health movement by circumventing described subject.

5. radiation image photography equipment according to claim 3, wherein, described rephotography control part increases the tightness of the fixture being used for fixing described subject.

6. radiation image photography equipment according to claim 1, wherein, described health mobility detect portion comprises:

Local motion-vector calculating part, it is for calculating at least one the local motion-vector of the local displacement represented in the overlapping region of described subject between described radiation image;

Health moves desired value calculating part, and it moves desired value for calculating health based on described local motion-vector; And

Health moves judging part, and it exists health move for moving desired value to judge whether based on described health.

7. radiation image photography equipment according to claim 6, wherein, described health moves the desired value that desired value calculating part calculates the translation amount of described subject and moves desired value as described health.

8. radiation image photography equipment according to claim 7, wherein, described health moves at least one also calculating in the desired value of the desired value of the three-dimensional amount of movement of described subject and the two-dimensional movement amount of described subject of desired value calculating part and moves desired value as described health.

9. radiation image photography equipment according to claim 1, wherein, described health mobility detect portion comprises the sensor of the movement for detecting described subject.