JP6514891B2 - Production area shooting system - Google Patents

Description

  The present invention relates to a production area photographing system for creating an overhead view image of a production area of a factory.

A production area of a factory that produces relatively heavy steel products of various shapes, structures and sizes is provided with a plurality of stages, and the plurality of stages are workpieces of various shapes, structures and sizes Are placed for assembly, welding, painting, inspection, etc.
In the side of these stages, a space for storing members and parts, a space for storing jigs and tools, etc. are arranged, and a passage for the passage of workers and carriages is also provided.

  When making a production plan, it is important to make effective use of the production area. For this purpose, it is necessary to sufficiently confirm the production area by visual observation and grasp the arrangement of works and parts, etc. and the progress of work, and then make a plan. When moving the work according to the progress of the work, it is also important to confirm the movement destination space to move the work, it is important to measure the vertical and horizontal dimensions of the movement destination space with a tape measure etc. There are many things to do. In any case, in order to understand the current status of the production area, staff members actually go to the production area, check visually, make notes on the layout of the work, etc. You must know the progress of the work.

  Patent Document 1 discloses a facility status management system in a factory. In this equipment status management system, various devices included in the layout information database storing information on the area configuration of the production area, the layout of the electrical equipment, compressed air equipment, manufacturing equipment etc., and the electrical equipment, compressed air equipment, manufacturing equipment etc. And an attribute information database storing attribute information relating to the layout of the electric equipment, for example, with the layout of the electric equipment superimposed on the layout of the area configuration and displayed on the display, the desired electric equipment in the electric equipment is designated. Attribute information can be displayed on the same screen. Thus, management of the position (place) of the equipment and the attribute information can be easily performed.

Unexamined-Japanese-Patent No. 2004-13197

  The facility status management system of Patent Document 1 prepares a database of various devices and devices of the facility and can easily display the position and attribute information of the devices and devices. It is not possible to know the position (place) and attribute information of various works being produced in the area, and various works whose place and condition change daily.

As described above, according to the prior art, the present conditions (position or place, work progress status, vacant space, horizontal and vertical dimensions of vacant space, etc.) of various works and parts being produced in a factory production area can be easily and accurately No system has been proposed that can create a graspable production area 俯瞰 image.
An object of the present invention is to provide a production area photographing system capable of creating a bird's-eye view image viewed from vertically above showing the current state of various works and parts being produced in a production area of a factory.

The production area photographing system according to claim 1 is a production area photographing system for creating a bird's-eye view image looking over the production area of a factory from above, wherein the predetermined height position above the production area can be moved in the longitudinal direction of the production area Producing a plurality of cut-out images respectively cut out from a plurality of images taken by a plurality of imaging means mounted on the body, a plurality of imaging means mounted on the moving body, and a plurality of imaging means each time the moving body moves a set distance or at set intervals. Image combining means for creating the overhead image by combining in the width direction and the longitudinal direction of the area, and when a plurality of cut-out images cut out from a plurality of images taken simultaneously are pasted on the image combining memory in consideration of the deviation of the shutter opening time when capturing by means characterized by including an a <br/> bonding position correcting means for correcting the bonding position of the plurality of cut out images There.

  According to a second aspect of the present invention, in the production area photographing system according to the first aspect of the present invention, the cutout area setting means for setting in advance characteristic extraction areas for cutting out cutout images from a plurality of images photographed by a plurality of imaging units. The image combining means combines a plurality of cut-out images respectively cut out based on the unique cut-out area from a plurality of images taken simultaneously by a plurality of image pickup means in the width direction to create a band-shaped wrinkle image It is characterized in that the bird's-eye view image is created by combining a plurality of band-like eyelid images in the longitudinal direction.

  A production area photographing system according to claim 3 is characterized in that, in the invention according to claim 1 or 2, a moving body position detecting means for detecting the position of the moving body in the movement direction in the production area is provided.

  According to a fourth aspect of the present invention, in the production area photographing system according to the third aspect, the movable body position detecting means comprises a plurality of light sensors having a light receiving function and equipped with the movable body, and a side of a production area. One or more reflecting plates attached in a different arrangement form for each pillar to a plurality of portions capable of facing the plurality of light sensors among a plurality of pillar members erected at predetermined intervals, and the movable body And a rotary encoder for detecting a rotation angle of a drive shaft for driving the drive wheel, wherein the movable body position detection means is configured to move the movable body based on light reception signals of the plurality of optical sensors and detection signals of the rotary encoder. It is characterized in that it is configured to detect the position of.

  According to the invention of claim 1, as described above, the moving body, the plurality of imaging means, and the plurality of images captured by the plurality of imaging means each time the moving body moves by the set distance or at the set time respectively Since the image combining means for combining the plurality of cutout images in the width direction and the longitudinal direction of the production area to create the overhead image is provided, the following effects can be obtained.

The plurality of imaging means can capture an image of the production area viewed from vertically above, so that it is possible to synthesize a plan view overhead view image of the production area viewed from vertically above the production area. Since the overhead view image is an image showing the current status of the work in the production area, parts, facilities, and tools, it should be used effectively for planning of the production plan, confirmation of the movement destination of the work, grasping of the empty space, etc. Can. Since the current situation of the production area can be grasped without going to the production area and observing, productivity of the production area can be enhanced.
And since the pasting position correction means was provided, the pasting position of a plurality of cutout images can be corrected with high accuracy to create a bird's eye image with high accuracy.

According to the second aspect of the invention, as described above, the image combining means is provided with the clipping area setting means capable of setting the unique clipping area in advance for each of the plurality of imaging means, and the image combining means A plurality of cut-out images respectively cut out based on the unique cut-out region are combined in the width direction to create a band-like eyelid image, and a plurality of the band-like eyelid images are combined in the longitudinal direction to create the bird's-eye view image .
Therefore, even when there is a variation in the accuracy of the mounting positions and orientations of a plurality of imaging means, a unique cutout area is set for each imaging means, and an image is synthesized using an image cut out from the inherent cutout area. The desired image synthesis can be performed, and a highly accurate bird's eye image can be created.

  According to the third aspect of the present invention, the moving body position detecting means for detecting the position of the moving body in the movement direction in the production area is provided, so that imaging can be performed by a plurality of imaging means each time the moving body moves by a set distance. it can. In addition, the position information of the moving object and the data of a plurality of images taken by a plurality of imaging means are stored in association, and the image synthesis is performed based on the information, so that the position of the production area in the longitudinal direction Consistency with the image can be ensured.

  According to the invention of claim 4, as described above, the movable body position detecting means comprises a plurality of light sensors mounted on the movable body and a plurality of pillars erected at predetermined intervals on the side of the production area. And a rotary encoder, and detects the position of the movable body based on the light reception signals of the plurality of optical sensors and the detection signals of the rotary encoder. It is possible to detect the position of the moving direction of the object with high accuracy.

It is a top view of the production area of the factory concerning an example of the present invention. It is a top view which shows the overhead traveling crane of a production area, several camera for imaging | photography, several imaging | photography area | regions, etc. FIG. It is a perspective view of a plurality of sets of optical sensors equipped on an overhead traveling crane. It is an explanatory view explaining a reflective board equipped near the upper end of # 1 pillar-#W pillar. It is a block diagram of the control system of a production area imaging | photography system. FIG. 6 is an explanatory view for explaining movement of a camera for shooting, a shooting interval, a shooting area, and a cutout image corresponding area. It is an explanatory view explaining arrangement of a plurality of photography cameras on an overhead crane, photography field, and a cutout picture corresponding field. It is explanatory drawing of the image of 1st camera, a cutting-out image, and a size. It is an explanatory view explaining technology which sets up a peculiar cutting out field. It is explanatory drawing of the image of # 1 camera, and its original cutting out area | region. It is explanatory drawing of the image of # 2 camera, and its original cutting out area | region. (A) A diagram for explaining the shift of the cutout image due to the shift of the shutter opening time, (b) an explanatory diagram in the case where the cutout image is synthesized ignoring the gap of the shutter opening time, (c) the shift of the shutter opening time It is explanatory drawing of the synthetic | combination image of the cutting-out image at the time of correct | amending an affixing position so as to correct. It is a flow chart of peculiar cutting out area setting control. It is a flowchart of the control program of a production area imaging | photography system. It is explanatory drawing which shows the principal part of the bird's-eye view image which made the cut-out image continue in multiple rows and multiple columns, and was synthesize | combined. It is a figure which shows the principal part of the bird's-eye view image synthesize | combined from the image which image | photographed the production area.

  Hereinafter, the form for carrying out the present invention is explained based on an example.

As shown in FIGS. 1 and 2, this production area photographing system 10 is a system for creating a bird's-eye view of a production area 1A of a factory 1 producing various products made of steel viewed from above (directly above) .
The size of the building 1 of this factory is width B and length L, and W columns 2 (# 1 to #W pillars) are erected at regular intervals on both sides of the building 1, and the upper ends of these columns 2 A girder member 3 covering the entire length is attached to the inner side in the vicinity, and a rail 4 for traveling the overhead crane is fixed to the upper surface of the girder member 3. In addition, a part of the longitudinal direction of the building 1 is a structure without a roof. Hereinafter, as shown in FIG. 1, north, south, east and west are defined and explained.

The overhead crane 5 (corresponding to a moving body, hereinafter referred to as a crane) mounted on the beam members 3 on both sides rolls on the rail 4 on the upper surface of the beam member 3, for example, a pair of driving wheels 5a and 1 It has a pair of driven wheels 5b, and can travel reciprocally in the longitudinal direction (east-west direction) of the building 1.
The traveling speed of the crane 5 is set by the speed setting lever, but can be switched to a rated speed of four steps of 1 to 4 notches, accelerates almost linearly immediately after the start of traveling, and almost linearly before stopping Slow down.

  As shown in FIG. 2, on the east side of the crane 5, N imaging cameras 6 (corresponding to imaging means, hereinafter referred to as “cameras”) are separated by a predetermined distance a in the width direction of the building in the common vertical plane V Everywhere, it is equipped at a predetermined height from the floor surface, and it is possible to shoot the full width of the production area 1A from directly above with N cameras 6. The camera 6 is a digital camera with a telephoto lens, and its resolution is capable of identifying an object of 10 cm × 10 cm on the floor surface.

  Each camera 6 is fixed to the main body frame 5c of the crane 5 by a camera fixing bracket 6a and a reinforcement brace 6b, and is precisely positioned so that the optical axis is directed downward in the vertical direction. The N cameras 6 described above are connected to a server unit 12 described later by a cable as shown in FIG. 5 and execute photographing based on a command from the server unit 12 and transmit the photographed image data to the server unit 12. It is configured to supply.

In FIG. 2, a hatched area Ar is a photographed area photographed by one camera 6, and one band-like area Br directly below the camera 6 is a cutout image B cut out from the image A of the area Ar. A cut-out image corresponding area corresponding to (see FIG. 8).
The images of the N cut-out image corresponding areas Br simultaneously photographed at one time by the N cameras 6 become band-like 俯瞰 images corresponding to the full width of the production area 1A. A plurality of sets of full-width band-wedge images created through multiple imaging while moving the crane 5 from the west end to the east end of the production area 1A are created over the entire length of the production area 1A, and they are synthesized Create a bird's-eye view of the entire production area with

  As shown in FIGS. 1 to 5, the crane position detection means 11 includes a rotary encoder 11A for detecting a rotation angle of a drive shaft 5d for driving the drive wheel 5a of the crane 5, and a plurality of sets of optical sensors mounted on the crane 5. 11a and one or more reflecting plates 11b respectively attached to # 1 column to #W column in different arrangement modes.

  The rotary encoder 11A has a rotary plate 11c in frictional contact with the drive shaft 5d, and an encoder main body 11d for detecting a rotation angle of the rotary plate 11c. The encoder main body 11d is mounted on the main body frame 5c via a mounting bracket 11e. It is fixed. A detection signal of the rotary encoder 11A is supplied to the server unit 12 by a cable.

  As shown in FIG. 4, one or more reflecting plates 11 b corresponding to a plurality of sets of optical sensors 11 a are arranged differently on each side 2 on the side of the # 1 column to the #W column on the production area 1A side (binary system The column number is detected from the combination of one or more received light signals received by the multiple sets of optical sensors 11a, and the columns numbered in order from the west end (# 1 column) of the building 1 The position of the pillar based on the west end of the production area 1A can be detected using the number.

  When detecting the position in the moving direction of the crane 5 in real time while moving the overhead crane 5 from the west end of the production area 1A to the east, a reflector is provided each time a light reception signal is input at the position of each column 2 The column number is detected from the arrangement mode of 11b (that is, the pattern of the light reception signal), and the position of each column 2 is detected from the column number, whereby the position in the traveling direction of the crane 5 (east and west based on # 1 column Detect the position of the direction). Then, the position of the crane 5 between each pillar 2 and the pillar 2 adjacent to the east thereof is calculated based on the detection signal of the rotary encoder 11A, and the current of the moving direction of the crane 5 based on the # 1 pillar The position can be detected with high accuracy in real time.

  Thus, the position in the moving direction of the crane 5 can be detected with high accuracy in real time based on the light reception signals of the plural sets of light sensors 11a and the detection signals of the rotary encoder 11A.

Next, a control system of the production area photographing system 10 will be described based on FIG.
The control system is equipped with a server unit 12 including a control unit. The server unit 12 is electrically connected to a plurality of sets of optical sensors 11a, N cameras 6 and a rotary encoder 11A. The unit 12 can also communicate wirelessly with the personal computer 13 and the tablet 14. The server unit 12 is installed at a predetermined installation place on the main body frame 5c of the overhead crane 5, and the personal computer 13 is installed at a predetermined place where the technical staff is stationed. The tablet 14 is mainly used for inputting a command for photographing.

The server unit 12 is a computer including a CPU, a main storage (ROM) and an auxiliary storage (RAM), a large capacity image data memory (which is composed of an SSD) for storing image data, a display screen and a keyboard , And an input / output interface. The main storage stores various control programs for executing the following arithmetic processing and control.
(A) Image shooting Acquisition and storage of image data (storage in image data memory)
・ Image data for setting specific cutting area (taken while the crane is stopped)
-Image data for creating a bird's-eye view image (shooting while moving the crane, entire production area)
(B) Setting of the unique extraction area, storage of area setting data (extraction area setting means)
(C) Synthesis of overhead view image (image synthesis means, paste position correction means)
・ Cutout image creation ・ Placement and pasting of clipped image (俯瞰 image creation), overhead image data storage (d) Crane position detection signal acquisition, crane position calculation (mobile body position detection means)
· Reading of light reception signal of optical sensor and detection signal of rotary encoder · Crane position calculation (e) Network of production area imaging system · Data transfer and control

  The image data memory stores at least image data to be used for setting the unique cutout area, and a cutout area storage unit for storing unique cutout area setting data, and a large amount of image data to be used for overhead image creation There is provided an image storage unit that stores the image together with the crane position and the imaging time, and a bird's-eye view image storage unit that stores the image data of the combined bird's-eye view image.

  The personal computer 13 is mainly used when reading out the bird's-eye view image created in the server unit 12 and referring to the bird's-eye view image on the screen.

  Next, the photographing area Ar and the cutout image corresponding area Br shown in FIG. 2 which are photographed by one camera 6 will be described. FIG. 6 is a diagram showing the dimension C in the longitudinal direction of the production area 1A of the imaging area Ar and the dimension b in the longitudinal direction of the production area 1A of the cutout image corresponding region Br. While traveling the crane 5 by acceleration → constant speed → deceleration, images are sequentially taken at every set time t. For example, positions after time t, 2t, 3t... Are positions T1, T2, T3. In addition, when the crane 5 travels at a constant speed, it corresponds to sequentially photographing each time the crane 5 moves a set distance b.

FIG. 7 is a diagram showing the plurality of cameras 6, the dimension D in the production area width direction of the imaging area Ar, and the dimension a in the production area width direction of the cutout image corresponding area Br.
Next, a technique for setting a unique cutout area corresponding to each camera 6 as a unique cutout area for cutting out the cutout image B from the image A obtained by photographing the imaging area Ar will be described based on FIGS. 8 and 9. .

  FIG. 8 shows an image A1 captured by the # 1 camera 6 and a cutout image B1 (which matches the unique cutout region) cut out from the image A1, and the width Ba of the cutout image B1 in the north-south direction is The width Bb corresponding to a on the production area 1A, and the length Bb in the east-west direction of the cutout image B1 is a length corresponding to b on the production area 1A. Since the cut-out image B1 is a long and thin rectangle, in the XY coordinate system of the image A1, a unique cut-out area can be set at three points P1, Q1, and R1.

Here, since there is an error in the mounting attitude (inclination with respect to the vertical, the rotation angle around the vertical axis) and the mounting position of the N cameras 6, it is possible to uniformly set the cutting out area from each captured image. It is not possible to do so, and it is necessary to set a unique cutout area associated with each camera 6. As shown in FIG. 9 using image data simultaneously captured by N cameras 6 in a state where the crane 5 is stopped at a predetermined place in the production area 1A after installation of the equipment of the production area imaging system 10 Set a unique extraction area.
However, at the time of this photographing, a standard scale M (this is a scale with a scale) shown separately in black and white shown in FIG. 9 is installed in a state of being accurately positioned in the north-south direction over the entire width of production area 1A, The image is photographed so that this reference scale M falls within the image.

FIG. 9 is an explanatory view of setting unique cutout areas B1 to B3 using photographed images A1 to A3 photographed by the cameras 6 of # 1 to # 3.
The photographed images A1 to A3 are displayed redundantly on the screen of the personal computer 13 so that the reference scales M are aligned in a straight line and the scale of the reference scale M is continuous. Arrange A3.

  In the example of FIG. 9, it is assumed that the X direction of the XY coordinate system of the image A1 is parallel to the reference scale M. Since the image A1 is an image on the start side, the vertex P1 of the unique clipping region B1 unique to the image A1 is a distance of 1/2 the width Ba of the north-south direction of the clipping image B1 from the center G1 of the image A1. Set to position. The vertex R1 is set to the position of the width Bb in the east-west direction of the cutout image B1 from the vertex P1.

  The vertex Q1 of the unique clipping region B1 (the vertex P2 of the unique clipping region B2 unique to the image A2) is set at a position corresponding to the middle between the center G1 and the center G2 of the image A2. The apex R2 of the unique cutout area B2 is set to the position of the width Bb in the east-west direction of the cutout image B1 from the apex Q1. Similarly, the vertex Q2 of the unique clipping region B2 (the vertex P3 of the unique clipping region B3 unique to the image A3) is set at a position corresponding to the middle between the center G2 and the center G3 of the image A3. The apex R3 of the unique cutout area B3 is set to the position of the width Bb in the east-west direction of the cutout image B1 from the apex Q2. Thereafter, by repeating similarly, unique cutout areas B4 to BN cut out from the image A4 to the image AN are set.

FIG. 10 shows an image A1, its inherent cutout area B1, and vertices P1, Q1, R1 of the image A1 in the XY coordinate system. FIG. 11 shows an image A2, its inherent cutout area B2, and XY coordinates of the image A2. It is a figure which shows vertices P2, Q2, and R2 in a system.
As described above, in the case where the cutout image Bi (i = 1, 2,... N) is cut out from the image Ai (i = 1, 2,... N), the characteristic cutout region defined in the XY coordinate system of each image Are cut out based on the vertices Pi, Qi, Ri (i = 1, 2,... N) of. Further, in the case where the cut out image Bi taken at the same time is pasted on the image combining memory, for example, the image is rotated and pasted by an inclination angle θi corresponding to the inclination angle θ2 with respect to the X axis shown in FIG.

Therefore, the following data is prepared and stored for creating a cutout image.
Camera number i (# 1 to #N), data of vertices Pi, Qi, Ri (i = 1, 2,... N) defining characteristic extraction areas, inclination angles θi (i = 1, 2,. ) Are stored in association with each other in the cutout area storage unit of the server unit 12. Note that the setting of the above-mentioned unique cutout area may be performed only once at the time of construction of the photographing system. The data of the photographed image Ai (i = 1, 2,... N) to be used for the above data creation is also stored in the cutout area storage unit of the server unit 12.

  Next, after preparing the data for cropping image creation, while making the crane 5 travel from the west end to the east end of the production area 1A, the production area 1A is repeatedly photographed simultaneously with N photographing cameras 6 to produce the production area Acquire the data of the photographed image of the entire area of 1A. This photographing is suitably performed, for example, before the start of operation of the factory or within a break time of lunch.

  Although imaging is performed at predetermined set time t intervals, imaging is performed at set time t intervals even during acceleration or deceleration, so the number of times of imaging while the crane 5 moves by the set distance b increases. Accordingly, the image can be synthesized with high accuracy by changing the width of Bb. However, the image data may be selected and adopted every time the crane 5 moves by the set distance b.

  Here, since the traveling speed is set by the manual operation of the crane driver at the time of acceleration or deceleration, it is difficult to calculate the moving distance of the crane 5 from the traveling time because the acceleration and the deceleration are not always constant. Also, since the driving wheel 5a may slip relative to the rail 4 when it rains, the crane position detection means 11 (mobile body position capable of detecting in real time the position of the crane 5 in the east-west direction (position in moving direction) (Corresponding to detection means), and stores the photographed image data in association with the position in the moving direction of the crane 5, and synthesizes the overhead image of the production area 1A based on the information.

A series of associated data for creating the overhead view image is as follows.
・ Data of camera number i ・ Data of photographed image Aij (including shutter open time data)
Movement Direction Position Data of Crane 5 Here, i is a subscript indicating a camera number, and i = 1, 2,. j is a subscript indicating the imaging order after the start of imaging, and for example, j = t, 2t, 3t. However, 1, 2, 3, ... may be adopted as j. Even when the crane 5 is in the acceleration region or the deceleration region, the image data captured at the set time t intervals is similarly stored.

  The moving direction position of the crane is calculated from time to time in the server unit 12 based on the light reception signal from the light sensor and the detection signal from the rotary encoder 11A, and the moving direction position at the time of photographing is stored as described above. Be done.

  By the way, when transmitting an imaging command simultaneously from the server unit 12 to the N cameras 6, a slight transmission delay occurs in the signal transmitted to the camera 6, so a slight deviation occurs at the shutter opening time of the camera 6. The shooting time (time of shutter opening) of the images A1, A2, A3,... Is shifted, and as shown in FIG. 12A, the shifted images B1, B2, B3,. In the illustrated example, the shutter opening times of the # 2 camera and the # 3 camera are delayed with respect to the # 1 camera.

  Therefore, assuming that the images A1, A2, A3,... Are taken at the same timing, as shown in FIG. 12B, pasting the cut out images B1, B2, B3,. It results in a composite image without image continuity in the direction. Therefore, as shown in FIG. 12C, the pasting positions of the cut out images B1, B2, B3,... Are corrected in consideration of the shift of the shutter opening time.

  When photographing the entire area of the production area 1A, the server unit 12 operates the N cameras 6 every set time t seconds based on the command from the tablet 14, and the momentary image received from the cameras 6 The data is stored in the image storage unit.

  The reference clock of the server unit 12 and the clock of each camera 6 are synchronized, and the time data of the time when the shutter is actually pressed in the corner camera 6 is transferred to the server unit 12 together with the image data.

In FIG. 12C, the shift amount α2 (pasting position correction amount) on the image in the crane movement direction of the cutout image B2 with respect to the cutout image B1 is as follows.
α2 = V × Δt2 × r
However, V is traveling speed at the time of photography of crane 5 Δt 2 delay time of photography time of # 2 camera with respect to # 1 camera r is a ratio (scale) to convert the actual size on the production area to the size on the image .
The cut-out images B3 to BN can also be calculated in the same manner as described above.

Here, the correction of the attachment position can also be performed in the following procedure.
(1) The server unit 12 stores in advance the relationship between the movement direction position of the crane 5 detected by the crane position detection means 11 and the time at the reference clock.
(2) When photographing with the camera 6, the time when the shutter is released together with the image is transferred to the server unit 12. At this time, the reference clock of the server unit 12 and the clock of each camera are synchronized, and the time of the reference clock is transferred.
(3) The movement direction position of the crane 5 at the time transferred in (2) is obtained from (1), and the position is adopted as the position of the camera 6 at the open time.

Next, a method of setting the unique cutout area will be supplementarily described with reference to FIG.
This method is implemented in the server unit 12. First, the crane 5 is stopped at a predetermined position (S1). Next, the # 1 to #N cameras are simultaneously shot, and the image data is stored in a predetermined area (storage unit) of the cutout area storage unit (S2). Next, the setting of the unique cutout area corresponding to the # 1 to #N cameras (the images taken by these cameras) is performed as described above (S3). Next, the data of the already described vertices Pi, Qi, Ri for which the unique cutout area has been set is stored in the cutout area storage unit (S4). This completes the setting of the unique extraction area. Note that the server unit 12 that executes S3 corresponds to "cutting area setting means".

Next, control for creating a bird's-eye view image will be supplementarily described based on FIG.
After the control is started, a process of initializing the system is performed (S10). Next, traveling of the crane 5 is started (S11). Next, it is determined whether or not it is the imaging timing for each set time t (S12), and when the determination is Yes, the detection signal of the encoder as the current position information of the crane 5 and the light reception signal from the optical sensor 11a are read. The position of the moving direction of the crane 5 is calculated (S13). If the determination in S12 is No, the process returns to S12.

  Next, in S14, control for simultaneously photographing with the # 1 to #N cameras is executed, and processing for storing the photographed image data in the image storage unit in association with the photographing time data and the crane position data is executed. Next, it is determined whether the imaging of the entire area of the production area 1A is completed (S15). If the determination is No, the process returns to S12, repeats S12 to S15, and when the determination of S15 is Yes, the process proceeds to S16. .

In S16, the traveling of the crane 5 is stopped. Next, in S17, composition of the overhead image is executed. In this case, the cutout image Bij across the entire production area 1A is first generated using the cutout image forming data described in paragraph [0043] and the data described in paragraph [0047]. However, i and j are the same as those described in paragraphs [0043] and [0047].
Next, in order from the smallest j, the process of combining the cutout images Bi (i = 1, 2,... N) for each j in the width direction is executed, and the band-like composite image Paste at the position corresponding to the crane position (position on the image composition memory). At this time, the sticking position is corrected in consideration of the shift of the shutter opening time. By repeating the above, it is possible to synthesize a bird's eye view of the entire production area 1A viewed from directly above.
The server unit 12 that executes S17 corresponds to "image combining means" including "pasting position correction means".

  FIG. 15 is an explanatory view for explaining a state in the course of attaching the band-shaped composite image to a position corresponding to the crane position. FIG. 16 is a schematic view showing the main part of the bird's-eye view image 1G. Next, the data of the bird's-eye view image 1G created in S17 is stored in the bird's-eye view image storage unit, and the data of the bird's-eye view image 1G is downloaded to the personal computer 13 based on the request from the personal computer 13 (S18). Thereafter, this control ends.

Next, the operation and effects of the production area photographing system 10 described above will be described.
A production area of a plurality of cut-out images respectively cut out from a plurality of images obtained by photographing the production area 1A with a plurality of cameras 6 every time the set distance travels by the overhead crane 5, a plurality of cameras 6 and the crane 5 Since the image combining means for combining the width direction and the longitudinal direction of 1A to create the overhead view image 1G is provided, the following effects can be obtained.

  The plurality of cameras 6 can capture an image in which the production area 1A is viewed vertically from above, so that a plan view overhead image 1G of the production area 1A in which the production area 1A is viewed vertically from above can be synthesized. Since the overhead image 1G is an image showing the current status of the work, parts, facilities, and tools in the production area 1A, it is effectively used for planning of the production plan, confirmation of the movement destination of the work, grasping of the empty space, etc. can do. Since the current situation of the production area 1A can be grasped without visiting the production area 1A and observing, productivity of the production area 1A can be enhanced.

As described above, a clipping region setting unit capable of setting unique clipping regions in advance for each of the plurality of cameras 6 is provided, and a plurality of clipping images respectively cut out based on the unique clipping regions from a plurality of images captured simultaneously by the plurality of cameras 6 The band-shaped eyelid image is created by combining in the width direction, and the bird's-eye view image 1G is created by combining a plurality of the band-shaped eyelid images in the longitudinal direction.
Therefore, even if there are variations in the accuracy of the mounting positions and orientations of the plurality of cameras 6, by setting a unique cutout area for each camera 6 and combining images using images cut out from the specific cutout areas It is possible to perform desired image synthesis, and it is possible to create a bird's eye image 1G with high accuracy.

  Since the crane position detection means for detecting the position in the movement direction of the crane 5 in the production area 1A is provided, the position information of the crane 5 can be detected with high accuracy. Further, the position information of the crane 6 and the data of a plurality of images taken by a plurality of cameras 6 are stored in association with each other, and the image synthesis is performed based on the information to obtain the position of the production area 1A in the longitudinal direction. The overhead image 1G can be matched.

  The moving body position detecting means is provided with a plurality of light sensors 11a mounted on the crane 5 and one or a plurality of different arrangement modes for each of a plurality of columns 2 erected at predetermined intervals on the side of the production area 1A. In order to detect the position of the crane 6 based on the light reception signals of the plurality of optical sensors 11a and the detection signals of the rotary encoder 11A, the position of the crane 5 in the moving direction is made high. The accuracy can be detected.

  When pasting a plurality of cut-out images cut out from a plurality of images taken at the same time on a memory for image composition, pasting of a plurality of cut-out images taking into account the shift of shutter open time when photographing with a plurality of cameras 6 Since the pasting position correction means for correcting the position is provided, it is possible to correct the pasting positions of the plurality of cut-out images with high accuracy and create a bird's eye image with high accuracy.

Next, an example in which the embodiment is partially changed will be described.
1) If there are cameras with different installation heights, create scaling data to match the scale of the image, store it along with the image data, store it, and set the scaling data before setting the unique cutout area The image is enlarged or reduced using this to set a unique clipping region. The enlargement / reduction data is stored together with the data for clipping image segmentation. Therefore, at the time of overhead image creation, the image is enlarged or reduced using the enlargement / reduction data and then the cutout image is cut out.

2) The number of cameras 6, the installation interval of the cameras 6, the installation height of the cameras, the width and length of the production area 1A, etc. are not limited to the examples described in the embodiments.
Although the case where imaging of the production area is performed every time the crane 5 moves a set distance (and every set time) has been described as an example, the configuration is such that "every time the set distance is moved" or "every set time". You may
3) The production area shooting system may be configured to shoot the entire production area or may be configured to shoot a part of the production area. In this case, while the crane 5 is moved to the place where the user wants to shoot, the crane 5 is started to travel, and shooting by the plurality of cameras 6 is started by the shooting command from the tablet 14. When imaging is stopped by the imaging stop command, the traveling of the crane 5 is stopped.

4) In the above embodiment, the setting process of the unique extraction area is automatically processed by the computer, but in the setting process of the unique extraction area, a plurality of images are displayed on the screen of the personal computer 13 and the images are moved or It is also possible to manually set the unique cutout area while moving it.
5) The production area to be photographed is not limited to heavy industry-related production areas, and the present invention can be applied to various production areas.

6) As a crane position detection means, a distance measurement means using laser light may be provided at one end side in the length direction of the production area, and a reflecting plate for reflecting the laser light may be provided on the crane.
7) Instead of the light sensor and reflector of the crane position detection means, attach a transmitter that emits a radio signal with a specific pattern for each pillar to each pillar, and receive a radio signal on the crane side You may equip it.
8) In addition, those skilled in the art can carry out various modifications to the above embodiment, and the present invention includes such modifications.

  The present invention provides a production area imaging system capable of creating a bird's-eye view image viewed from vertically above showing the current status of various works and parts being produced in a factory production area.

1A Production area 1G 俯瞰 Image 5 Overhead crane (mobile)
6 Camera for shooting (imaging means)
DESCRIPTION OF SYMBOLS 10 Production area imaging system 11 Crane position detection means 11a Optical sensor 11b Reflective plate 11A Rotary encoder 12 Server unit (cutting out area setting means, image combining means)
S3 clipping region setting means S17 image combining means Ai photographed image Bi clipping image (specific clipping region)

Claims (4)

In a production area imaging system that creates a bird's-eye view of the factory's production area from above,
A movable body capable of moving a predetermined height position above the production area in the longitudinal direction of the production area;
A plurality of imaging means mounted on the moving body;
A plurality of cut-out images respectively cut out from a plurality of images taken by a plurality of image pickup means each time the moving body moves by a set distance or at set time intervals are combined in the width direction and the longitudinal direction of the production area to create the overhead view image Image combining means,
When pasting a plurality of cut-out images cut out from a plurality of images taken at the same time on a memory for image composition, pasting a plurality of cut-out images taking into account the shift of shutter open time at the time of photographing with a plurality of imaging means Paste position correction means for correcting the position;
A production area photographing system characterized by comprising. A cutout area setting unit configured to set in advance, for each of the imaging units, a unique cutout area for respectively cutting out a cutout image from a plurality of images photographed by a plurality of imaging units;
The image combining means combines a plurality of cut-out images respectively cut out based on the unique cut-out area from a plurality of images taken simultaneously by a plurality of image pickup means in the width direction to create a band-like scale image The production area photographing system according to claim 1, wherein the bird's-eye view image is created by combining a plurality of images in the longitudinal direction.   The production area photographing system according to claim 1 or 2, further comprising a movable body position detection means for detecting the position of the movable body in the movement direction in the production area. The movable body position detection means includes a plurality of light sensors having a light projection / reception function equipped on the movable body, and a plurality of the plurality of pillar members erected at predetermined intervals on the side of the production area. And one or more reflectors attached in a different arrangement for each column at a plurality of locations that can be opposed to the light sensor, and a rotary encoder for detecting the rotation angle of the drive shaft that drives the drive wheel of the movable body. Equipped
The mobile object position detection means is configured to detect the position of the mobile object based on light reception signals of the plurality of optical sensors and detection signals of the rotary encoder. Production area shooting system described.