JP5662088B2 - Remote visual inspection device - Google Patents

Description

  The present invention relates to a remote visual inspection apparatus.

  In a nuclear power plant, a visual inspection of a structure in the nuclear reactor is regularly performed in order to confirm the soundness of the nuclear reactor. In the visual inspection, in order to prevent the inspector from being exposed, indirect visual inspection is performed instead of direct visual inspection. In indirect visual inspection, an inspector remotely operates an imaging device such as an underwater television camera, and inspects for the presence or absence of defects on the surface of the inspection target based on an image captured by the imaging device displayed on the display device.

  Conventionally, in remote operation of an imaging apparatus used for indirect viewing, an operator manually operates a wire, an operation pole, or the like according to an inspector's instruction. Therefore, it is difficult to control the camera when the imaging device is inserted into a narrow portion. As a result, even if the inspector wants the imaging device to approach the inspection target, the worker cannot sufficiently bring the imaging device close to the inspection target. As a result, it is not possible to obtain an image taken from a viewpoint necessary for visual inspection.

  It is also difficult to make the imaging device stationary. Therefore, a feeling of camera shake occurs in the video imaged by the imaging device. As a result, images taken from the viewpoint necessary for visual inspection cannot be obtained, and the eyes of the inspector are fatigued.

  In order to solve these problems, Patent Document 1 discloses a technique for creating a high-resolution image having a higher resolution than a video image captured by an imaging device from a time-series image captured by the imaging device. Yes.

  However, in the technique of Patent Document 1, image processing for a plurality of images is necessary to create a high-resolution image. Therefore, the time for image processing becomes longer. As a result, there is a problem that the video cannot be projected on the display device in real time, the video projected on the display device becomes intermittent, and the inspector feels uncomfortable.

JP 2009-103682 A

  The present invention has been made to solve the above-described problem, and obtains an image taken from a viewpoint necessary for visual inspection, eliminates a feeling of camera shake of an image taken by an imaging device, and high resolution. An object of the present invention is to provide a remote visual inspection device for displaying an image on a display device in real time using an image.

According to this embodiment,
An imaging unit that captures an image of the surface of the inspection object while moving;
A movement amount calculation unit that calculates a movement amount that is a change amount of the position and orientation of the imaging unit based on a difference between feature amounts of at least two image data constituting the video;
A movement amount memory for storing the movement amount history;
Based on the history of the movement amount, a position estimation unit that estimates a suitable position of the imaging unit and a posture suitable for visual inspection for an inspector;
There is provided a remote visual inspection apparatus comprising: an image conversion unit that generates processed image data corresponding to the preferred position by performing geometric transformation on the image data based on the preferred position.

  According to the present invention, it is possible to obtain an image captured from a viewpoint necessary for visual inspection, to eliminate the hand shake feeling of the image captured by the imaging device, and to display the image on a display device using a high resolution image. Can be projected in real time.

The block diagram which shows the structure of the remote visual inspection apparatus 1 which concerns on this embodiment. 1 is a block diagram showing a configuration of an image processing apparatus 10 according to a first embodiment. The figure which shows the data structure of the log | history of a movement amount. The flowchart which shows the procedure of the remote visual inspection process which concerns on 1st Embodiment. Schematic for demonstrating the remote visual inspection process of FIG. The block diagram which shows the structure of the image processing apparatus 10 which concerns on 2nd Embodiment. The flowchart which shows the procedure of the remote visual inspection process which concerns on 2nd Embodiment. The flowchart which shows the procedure of the image selection (S711) of FIG. The block diagram which shows the structure of the image processing apparatus 10 which concerns on 3rd Embodiment. The flowchart which shows the procedure of the remote visual inspection process which concerns on 3rd Embodiment. The figure for demonstrating the high resolution image generation (S1006) of FIG. The block diagram which shows the structure of the image processing apparatus 10 which concerns on 4th Embodiment. The flowchart which shows the procedure of the remote visual inspection process which concerns on 4th Embodiment. The block diagram which shows the structure of the image processing apparatus 10 which concerns on 5th Embodiment. The block diagram which shows the structure of the image processing apparatus 10 which concerns on 6th Embodiment. The figure which shows the data structure of prediction information. The flowchart which shows the procedure of the prediction process which concerns on 6th Embodiment.

  Hereinafter, the present embodiment will be described in detail with reference to the drawings.

  A remote visual inspection apparatus according to this embodiment will be described. FIG. 1 is a block diagram showing a configuration of a remote visual inspection apparatus 1 according to this embodiment.

  As shown in FIG. 1, the remote visual inspection device 1 according to the first embodiment includes an image processing device 10, a remote operation device 20, an imaging device 30, a display device 40, and an interface 50. It is connected to a network 60 such as (Local Area Network).

  The remote operation device 20 is a device that operates the imaging device 20. The remote control device 20 is connected to the network 60. When the worker operates the remote control device 20 according to the instructor's instruction, the remote control device 20 generates a control signal CTR based on the operation of the worker. Next, the remote operation device 20 transmits a control signal CTR to the imaging device 30.

The imaging device 30 is a device that captures an image of the surface of the inspection target of the power plant while moving in the power plant. The imaging device 30 receives the control signal CTR transmitted from the remote control device 20. Next, the imaging device 30 captures an image to be inspected while moving according to the control signal CTR. Next, the imaging device 30 generates a plurality of pieces of image data IMG _{0 to} IMG _n constituting a plurality of frames F _{0 to} F _n (n is an integer of 1 or more) of the captured video. Next, the imaging device 30 supplies a plurality of image data IMG _{0 to} IMG _n to the image processing device 10. For example, the imaging device 30 is an underwater television camera, is supported by a wire, a pole, or the like, and takes a plurality of pieces of image data IMG _{0 to} IMG by capturing an image of the surface of the inspection target when approaching the vicinity of the inspection target. _n is generated.

  The interface 50 is a module that receives setting information CONF for the image processing apparatus 10 and transmits data D (hereinafter referred to as “processing data”) D of the processing result of the image processing apparatus 10. The interface 50 receives the setting information CONF. Next, the interface 50 supplies the setting information CONF to the image processing apparatus 10. Further, the interface 50 receives the processing data D from the image processing apparatus 10. Next, the interface 50 transmits the processing data D. For example, the interface 50 may receive the setting information CONF from an input device (not shown) such as a keyboard or from a network terminal (not shown) connected via the network 60. Further, the interface 50 may transmit the processing data D to the remote control device 20 or to a network terminal.

The image processing apparatus 10 is an apparatus that performs image processing on the image data IMG. The image processing device 10 receives a plurality of image data IMG _{0 to} IMG _n from the imaging device 30. Then, the image processing apparatus 10, by performing image processing on a plurality of image data _IMG 0 _{~IMG n,} to generate a plurality of processed image data IMG' ₀ ~IMG' _n. Next, the image processing apparatus 10 supplies a plurality of processed image data IMG ′ _{0 to} IMG ′ _n to the display device 40 and also supplies processed data D to the interface 50.

The display device 40 is a device that displays an image. The display device 40 receives a plurality of processed image data IMG ′ _{0 to} IMG ′ _n from the image processing device 10. Next, the display device 40 displays the image of the surface to be inspected by reproducing the frames F _{0 to} F _n configured by the plurality of processed image data IMG ′ _{0 to} IMG ′ _n . For example, the display device 40 is an LCD (Liquid Crystal Display).

Hereinafter, the image processing apparatus 10, by performing image processing on a plurality of image data _IMG 0 _{~IMG n,} describes an example of generating a plurality of processed image data IMG' ₀ ~IMG' _n. Hereinafter, image data (hereinafter, referred to as “target image data”) to be processed by the image processing apparatus 10 is represented as “IMG _i (i is an integer of 0 to n)”.

(First embodiment)
A first embodiment will be described. The first embodiment is an example in which a suitable position is estimated based on the movement amount history of the imaging apparatus, and geometric transformation is performed on the image data based on the estimated suitable position.

  The image processing apparatus 10 according to the first embodiment will be described. FIG. 2 is a block diagram illustrating a configuration of the image processing apparatus 10 according to the first embodiment.

  As illustrated in FIG. 2, the image processing apparatus 10 according to the first embodiment includes a movement amount calculation unit 11, a movement amount memory 12, a position estimation unit 13, and an image conversion unit 14.

The movement amount calculation unit 11 is a module that calculates the amount of change in the position and orientation of the imaging device 30 (hereinafter referred to as “movement amount”). The movement amount calculation unit 11 receives a plurality of pieces of image data IMG _{i to} IMG _n from the imaging device 30 and receives setting information CONF (algorithm) from the interface 50. Then, the movement amount calculating unit 11 uses the setting information CONF (algorithm), the feature amount _{S i} of the target image data IMG _i, before the frame _{F i} formed by the subject image data IMG _i frames (hereinafter, A feature quantity S _{i-k of} image data (hereinafter referred to as “reference image data”) IMG _i-k constituting F _i-k (k is an integer of 1 to n)) is compared with “previous frame”. Thus, the movement amount of the imaging device 30 (that is, the position and orientation of the imaging device 30 when the reference image data IMG _i-k is generated and the position and orientation of the imaging device 30 when the target image data _i is generated). Calculate the difference. In other words, the movement amount calculation unit 11 calculates the movement amount of the imaging device 30 based on the difference between the feature amounts of at least two image data constituting the video imaged by the imaging device 30. Then, the movement amount calculating section 11 generates movement amount information M _i representing the amount of movement. Next, the movement amount calculation unit 11 stores the movement amount information M _i in the movement amount memory 12.

Movement amount memory 12 is a module for storing the movement amount information M _i. That is, the movement amount memory 12 stores a movement amount history. As shown in FIG. 3, the movement amount history is a combination of image data IMG _i and movement amount information M _i . The movement amount information M _i includes a position change amount L _i that represents a change amount of the position of the imaging device 30 when the target image data IMG _i is generated, and a posture change amount P _i that represents a change amount of the posture of the imaging device 30. .

The position estimation unit 13 is a module that estimates the position and orientation of the imaging device 30 suitable for visual inspection for the inspector (hereinafter referred to as “preferred position”). The position estimation unit 13 estimates a suitable position based on the movement amount history stored in the movement amount memory 12. Next, the position estimation unit 13 generates suitable position information PP _i representing a suitable position. Next, the position estimation unit 13 supplies the suitable position information PP _i to the image conversion unit 14.

The image conversion unit 14 is a module that performs geometric conversion on the target image data IMG _i . The image conversion unit 14 receives a plurality of image data IMG _{0 to} IMG _n from the imaging device 30, receives suitable position information PP _i from the position estimation unit 13, and receives setting information CONF (algorithm) from the interface 50. Next, the image conversion unit 14 performs the geometric conversion on the target image data IMG _i using the setting information CONF (algorithm) to generate the processed image data IMG ′ _i corresponding to the suitable position information PP _i . By repeatedly generating the processed image data IMG ′ _i , a plurality of processed image data IMG ′ _{0 to} IMG ′ _n is generated. Next, the image conversion unit 14 supplies a plurality of processed image data IMG ′ _{0 to} IMG ′ _n to the display device 40.

  The operation of the remote visual inspection apparatus 1 according to the first embodiment will be described. FIG. 4 is a flowchart showing a procedure of remote visual inspection processing according to the first embodiment. FIG. 5 is a schematic diagram for explaining the remote visual inspection processing of FIG.

  The remote visual inspection process in FIG. 4 starts when the inspector gives an instruction to the operator to operate the remote control device 20.

  <FIG. 4: Remote Operation (S401)> The remote operation device 20 generates a control signal CTR based on the operation of the worker. Next, the remote operation device 20 transmits a control signal CTR to the imaging device 30. Thereby, the imaging device 30 is remotely operated.

<FIG. 4: Imaging (S402)> The imaging apparatus 30 captures an image to be inspected while moving according to the control signal CTR. Then, the imaging device 30 generates a plurality of image data _IMG 0 _{~IMG n} constituting a plurality of frames _F 0 to F _n of the photographed image. Next, the imaging device 30 supplies a plurality of image data IMG _{0 to} IMG _n to the image processing device 10.

<FIG. 4: Movement Amount Calculation (S403)> The movement amount calculation unit 11 uses the setting information CONF (algorithm) and the feature amount S _i of the target image data IMG _{i and} the feature amount S of the reference image data IMG _{i-k. The} amount of movement of the imaging device 30 is calculated by comparing _i−k . Then, the movement amount calculating section 11 generates movement amount information M _i representing the amount of movement. Next, the movement amount calculation unit 11 stores the movement amount information M _i in the movement amount memory 12. As a result, the movement amount history is stored in the movement amount memory 12 as shown in FIG.

For example, when the imaging device 30 simply translates (that is, when the position of the imaging device 30 changes and the orientation of the imaging device 30 does not change), the movement amount calculation unit 11 uses the reference image data IMG. The imaging device 30 when the reference image data IMG _i-k is generated and the imaging device 30 when the target image data IMG _i is generated by the block matching method using the luminance of _{i-k and} the luminance of the target image data IMG _i. Calculate the relative movement between and.

For example, when the imaging device 30 rotates and moves (that is, when the position and orientation of the imaging device 30 change), the movement amount calculation unit 11 uses the Harris operator to characterize the reference image data IMG _i-k . The point S _i-k and the feature point S _i of the target image data IMG _i are extracted. Next, the movement amount calculation unit 11 uses the Lanzac method to generate the target image data IMG for the imaging device 30 when the reference image data IMG _i-k is generated based on the two feature points S _i-k , S _{i. A} homography matrix expressing the amount of movement of the imaging device 30 when _i is generated is generated. Next, the movement amount calculation unit 11 calculates the movement amount of the imaging device 30 by associating the feature points S _i-k and the feature points S _i with a homography matrix.

<FIG. 4: Preferential Position Estimation (S <b>404)> The position estimation unit 13 estimates a preferable position based on the movement amount history stored in the movement amount memory 12. Next, the position estimation unit 13 generates suitable position information PP _i representing a suitable position. Next, the position estimation unit 13 supplies the suitable position information PP _i to the image conversion unit 14. For example, the suitable position information PP _i is information representing a state where the imaging device 30 is stationary or moving at a constant speed at a low speed.

<FIG. 4: Image Conversion (S405)> The image conversion unit 14 uses the setting information CONF (algorithm) to perform geometric conversion on the target image data IMG _i based on the preferred position information PP _i , thereby processing image data. IMG ′ _i is generated. For example, the geometric transformation is an affine transformation. By repeatedly generating the processed image data IMG ′ _i , a plurality of processed image data IMG ′ _{0 to} IMG ′ _n is generated. Next, the image conversion unit 14 supplies a plurality of processed image data IMG ′ _{0 to} IMG ′ _n to the display device 40.

For example, when the worker is not operating the remote control device 20, as shown in FIG. 5, the image conversion unit 14 determines the position of the feature point S _i-k of the reference image data IMG _i-k and the target image. Geometric transformation is performed on the target image data IMG _i so that the position of the feature point S _i of the data IMG _i is substantially equal. As a result, an image equivalent to an image captured while the imaging device 30 is stationary can be obtained.

For example, when a worker is operating the remote control device 20, the image conversion unit 14 estimates the motion model of the imaging device 30 as a constant velocity motion model based on the control signal CTR, and the estimated constant velocity motion. Processed image data IMG ′ _i is generated by performing geometric transformation on the target image data IMG _i according to the model. Specifically, the image converting unit 14, as a feature point S _i of the target image data IMG _i are the same as manner that uniform motion to or homography matrix is time, subjected to geometric transformation on the target image data IMG _i . Thereby, an image equivalent to an image captured in a state where the imaging device 30 is moving at a constant speed at a low speed is obtained.

<Figure 4: Display (S406)> display device 40 by playing the frame _F 0 to F _n composed of a plurality of processed image data IMG' ₀ ~IMG' _n, and displays an image. When the display (S406) ends, the remote visual inspection process of FIG. 4 ends.

According to the first embodiment, the position estimation unit 13 estimates the preferred position information PP _i based on the movement amount history, and the image conversion unit 14 performs geometric transformation on the image data IMG based on the preferred position information PP _i. . As a result, it is possible to obtain an image taken from a viewpoint necessary for visual inspection and having little camera shake. As a result, the accuracy of visual inspection can be improved, and the eye fatigue of the inspector can be reduced. Furthermore, even if the operator does not operate the remote control device 20 so that the imaging device 30 is stationary, it is possible to obtain a video image equivalent to a video imaged while the imaging device 30 is stationary. As a result, the burden on the worker can be reduced and the time required for the remote visual inspection can be shortened.

(Second Embodiment)
A second embodiment will be described. The second embodiment is an example of an image processing apparatus in which an image sorting unit that sorts image data that meets the sorting conditions determined by the inspector is added as compared to the image processing apparatus according to the first embodiment. . In addition, the description similar to said embodiment is abbreviate | omitted.

  An image processing apparatus 10 according to the second embodiment will be described. FIG. 6 is a block diagram illustrating a configuration of the image processing apparatus 10 according to the second embodiment.

  As illustrated in FIG. 6, the image processing apparatus 10 according to the second embodiment includes a movement amount calculation unit 11, a movement amount memory 12, a position estimation unit 13, an image conversion unit 14, an image selection unit 15, And an image memory 16. The movement amount memory 12 and the position estimation unit 13 are the same as those in the first embodiment.

The movement amount calculation unit 11 is a module that calculates the movement amount of the imaging device 30. The movement amount calculation unit 11 receives a plurality of pieces of image data IMG _{i to} IMG _n from the imaging device 30 and receives setting information CONF (algorithm) from the interface 50. Then, the movement amount calculating unit 11 uses the setting information CONF (algorithm) compares the feature quantity _{S i} of the target image data IMG _i, a feature amount _{S i-k} of the reference image data _{IMG i-k,} a Thus, the amount of movement of the imaging device 30 is calculated. In other words, the movement amount calculation unit 11 calculates the movement amount of the imaging device 30 based on the difference between the feature amounts of at least two image data constituting the video imaged by the imaging device 30. Then, the movement amount calculating section 11 generates movement amount information M _i indicating the amount of movement. Then, the movement amount calculating unit 11, the movement amount information M _i, and stores the movement amount memory 12, and supplies the image selecting unit 15.

The image sorting unit 15 is a module that sorts the target image data IMG _i . The image selection unit 15 receives the movement amount M _i from the movement amount calculation unit 11, receives the target image data IMG _i from the imaging device 30, and receives setting information CONF (selection condition) from the interface 50. Next, the image selection unit 15 determines whether the target image data IMG _i meets the setting information CONF (selection condition) based on the movement amount M _i . Next, the image sorting unit 15 stores the target image data IMG _i that conforms to the setting information CONF (sorting condition) in the image memory 16 and discards the target image data IMG _i that does not conform to the setting information CONF (sorting condition). For example, the setting information CONF (selection condition) is information arbitrarily determined by the inspector, and “the field of view of the frame F _i constituted by the target image data IMG _{i is} constituted by the reference image data IMG _i-k. It is different from the field of view of the front frame F _i-k ”. Specifically, the setting information CONF (selection condition) includes values such as contrast, brightness, and range. As a result, the image memory 16 stores image data necessary and sufficient for videos with different visual fields.

The image memory 16 is a module that stores the target image data IMG _i selected by the image selection unit 15. That is, the image memory 16 stores target image data IMG _i that meets conditions (for example, contrast, brightness, and range) desired by the inspector.

The image conversion unit 14 is a module that performs geometric conversion on the target image data IMG _i . The image conversion unit 14 receives the preferred position information PP _i from the position estimation unit 13 and also receives the setting information CONF (algorithm) from the interface 50. Next, the image conversion unit 14 reads a plurality of target image data IMG _{0 to} IMG _n from the image memory 16, and performs geometric conversion on the target image data IMG _i using the setting information CONF (algorithm). Processed image data IMG ′ _i corresponding to the information PP _i is generated. By repeatedly generating the processed image data IMG ′ _i , a plurality of processed image data IMG ′ _{0 to} IMG ′ _n is generated. Next, the image conversion unit 14 supplies a plurality of processed image data IMG ′ _{0 to} IMG ′ _n to the display device 40.

  The operation of the remote visual inspection apparatus 1 according to the second embodiment will be described. FIG. 7 is a flowchart showing a procedure of remote visual inspection processing according to the second embodiment. FIG. 8 is a flowchart showing the procedure of image selection (S711) in FIG.

  The remote visual inspection process in FIG. 7 starts when the inspector gives an instruction to the operator to operate the remote control device 20.

  <FIG. 7: Remote Operation (S701) and Imaging (S702)> The same as in the first embodiment.

<FIG. 7: Movement Amount Calculation (S703)> The movement amount calculation unit 11 uses the setting information CONF (algorithm) and the feature amount S _i of the target image data IMG _{i and} the feature amount S of the reference image data IMG _{i-k. The} amount of movement of the imaging device 30 is calculated by comparing _i−k . Then, the movement amount calculating section 11 generates movement amount information M _i representing the amount of movement. Then, the movement amount calculating unit 11, the movement amount information M _i indicating the calculated movement amount, and stores the movement amount memory 12, and supplies the image selecting unit 15. As a result, as shown in FIG. 3, the movement amount history of the imaging device 30 is stored in the movement amount memory 12, and the movement amount history is supplied to the image selection unit 15.

  <FIG. 7: Preferential Position Estimation (S704)> The same as in the first embodiment.

<FIG. 7: Image Sorting (S711)> The image sorting unit 15 determines whether the target image data IMG _i meets the setting information CONF (sorting condition) based on the movement amount M _i . For example, the image selection (S711) is executed according to the procedure of FIG.

<Figure 8: S801> image selecting unit 15 compares the threshold Mth amount of movement _{M i} and the movement amount. The movement amount threshold value Mth is included in the setting information CONF (selection condition). When the movement amount M _i is larger than the threshold Mth (that is, when the field of view of the frame F _i configured by the target image data IMG _i is different from the field of view of the previous frame F _i-k ) (S801—YES), Data storage (S802) is executed. When the movement amount M _i is equal to or less than the threshold value Mth (that is, when the field of view of the frame F _i configured by the target image data IMG _i is the same as the field of view of the previous frame F _i-k ) (S801-NO), Image data discarding (S811) is executed.

<FIG. 8: Target Image Data Storage (S802)> The image selection unit 15 stores the target image data IMG _i in the image memory 16. When the target image data storage (S802) ends, the image selection in FIG. 8 ends.

<FIG. 8: Discarding Target Image Data (S811)> The image sorting unit 15 discards the target image data IMG _i . When the target image data discard (S811) ends, the image selection in FIG. 8 ends.

<FIG. 7: Image Conversion (S705)> The image conversion unit 14 reads the target image data IMG _i from the image memory 16, and uses the setting information CONF (algorithm) to determine the target image data IMG based on the preferred position information PP _i. by performing geometric conversion to _i, to generate the processed image data IMG' _i. By repeatedly generating the processed image data IMG ′ _i , a plurality of processed image data IMG ′ _{0 to} IMG ′ _n is generated. Next, the image conversion unit 14 supplies a plurality of processed image data IMG ′ _{0 to} IMG ′ _n to the display device 40.

  <FIG. 7: Display (S706)> The same as in the first embodiment. When the display (S706) ends, the remote visual inspection process of FIG. 7 ends.

According to the second embodiment, the image sorting unit 15 stores the target image data IMG _i that conforms to the setting information CONF (sorting condition) in the image memory 16, and the target image data that does not conform to the setting information CONF (sorting condition). Discard IMG _i . As a result, even when the amount of movement of the imaging device 30 is increased due to, for example, the hand shake feeling of the image captured by the imaging device 30, an image close to the size of the field of view of the imaging device 30 can be obtained. As a result, the efficiency of visual inspection is improved.

(Third embodiment)
A third embodiment will be described. The third embodiment is an example of an image processing apparatus to which a high resolution image generation unit that generates high resolution image data is added as compared with the image processing apparatus according to the second embodiment. In addition, the description similar to said embodiment is abbreviate | omitted.

  An image processing apparatus 10 according to the third embodiment will be described. FIG. 9 is a block diagram illustrating a configuration of the image processing apparatus 10 according to the third embodiment.

  As illustrated in FIG. 9, the image processing apparatus 10 according to the third embodiment includes a movement amount calculation unit 11, a movement amount memory 12, a position estimation unit 13, an image conversion unit 14, an image selection unit 15, An image memory 16 and a high-resolution image generation unit 17 are provided. The movement amount calculation unit 11, the movement amount memory 12, the position estimation unit 13, the image selection unit 15, and the image memory 16 are the same as those in the second embodiment.

The image conversion unit 14 is a module that performs geometric conversion on the target image data IMG _i . The image conversion unit 14 receives the preferred position information PP _i from the position estimation unit 13 and also receives the setting information CONF (algorithm) from the interface 50. Next, the image conversion unit 14 reads a plurality of target image data IMG _{0 to} IMG _n from the image memory 16, and performs geometric conversion on the target image data IMG _i using the setting information CONF (algorithm). Processed image data IMG ′ _i corresponding to the information PP _i is generated. By repeatedly generating the processed image data IMG ′ _i , a plurality of processed image data IMG ′ _{0 to} IMG ′ _n is generated. Next, the image conversion unit 14 supplies the plurality of processed image data IMG ′ _{0 to} IMG ′ _n to the high-resolution image generation unit 17 and the display device 40.

The high resolution image generation unit 17 is a module that generates high resolution image data HIMG _i . The high resolution image data HIMG _i has a higher resolution than the resolution of the processed image data IMG ′ _i . The high resolution image generation unit 17 receives the processing image data IMG ′ _i from the image conversion unit 14 and also receives setting information CONF (algorithm) from the interface 50. Then, the high resolution image generation unit 17 reads at least one target image data IMG _i from the image memory 16, using the setting information CONF (algorithm), combines the processed image data IMG' _i and the target image data IMG _i As a result, high-resolution image data HIMG _i is generated. By repeatedly generating the high-resolution image data HIMG _i , a plurality of high-resolution image data HIMG _{0 to} HIMG _n is generated. Next, the high resolution image generation unit 17 supplies the plurality of high resolution image data HIMG _{0 to} HIMG _n to the display device 40.

The display device 40 receives a plurality of processed image data IMG ′ _{0 to} IMG ′ _n and a plurality of high-resolution image data HIMG _{0 to} HIMG _n from the image processing device 10. Next, the display device 40 reproduces the frames F _{0 to} F _n configured by the plurality of processed image data IMG ′ _{0 to} IMG ′ _n to display a normal resolution video and a plurality of high resolution image data. display high-resolution images by HIMG ₀ ~HIMG _n by reproducing the formed high-resolution frame _HF 0 _{~HF n.} As a result, a normal resolution video and a high resolution video are simultaneously displayed on the display device 40.

  The operation of the remote visual inspection apparatus 1 according to the third embodiment will be described. FIG. 10 is a flowchart illustrating a procedure of remote visual inspection processing according to the third embodiment.

  The remote visual inspection process in FIG. 10 starts when the inspector gives an instruction to the operator to operate the remote control device 20.

  <FIG. 10: Remote Operation (S1001) to Predicting Preferred Position (S1004) and Image Sorting (S1011)> The same as in the second embodiment.

<FIG. 10: Image Conversion (S1005)> The image conversion unit 14 reads the target image data IMG _i from the image memory 16, and uses the setting information CONF (algorithm) to determine the target image data IMG based on the preferred position information PP _i. by performing geometric conversion to _i, to generate the processed image data IMG' _i. By repeatedly generating the processed image data IMG ′ _i , a plurality of processed image data IMG ′ _{0 to} IMG ′ _n is generated. Next, the image conversion unit 14 supplies the plurality of processed image data IMG ′ _{0 to} IMG ′ _n to the high-resolution image generation unit 17 and the display device 40.

<FIG. 10: High-Resolution Image Generation (S1006)> The high-resolution image generation unit 17 reads at least one target image data IMG _i from the image memory 16, and uses the setting information CONF (algorithm) to process the processed image data IMG ′. High resolution image data HIMG _i is generated by combining _i and target image data IMG _i . By repeatedly generating the high-resolution image data HIMG _i , a plurality of high-resolution image data HIMG _{0 to} HIMG _n is generated. Next, the high resolution image generation unit 17 supplies the plurality of high resolution image data HIMG _{0 to} HIMG _n to the display device 40.

For example, as shown in FIG. 11, the high resolution image generation unit 17, using the technique of parabola fitting, etc., a frame _F'i composed of processed image data IMG' _i, is constructed by the target image data IMG _i a frame _{F i} that, as overlap synthesizes the processing image data IMG' _i and the target image data IMG _i. The frame F ′ _i includes a specific pixel PEL ′ _i . The frame F _i includes a specific pixel PEL _i . As a result, the portion where the specific pixel PEL ′ _i and the specific pixel PEL _i overlap becomes the high resolution pixel HPEL _i . The value of the high resolution pixels HPEL _i is the average value of a particular pixel PEL _i to the value of a particular pixel PEL' _i, is two times the value of a particular pixel PEL _i. That is, the resolution of the high resolution pixel HPEL _i is twice the resolution of the specific pixel PEL ′ _i . Thus, constructed high-resolution frame HF _i by the high resolution image data Himg _i is obtained. The high resolution frame HF _i includes high resolution pixels HPEL _i .

The high resolution image generation unit 17, the processed image data IMG' _i, a plurality of target image data _{IMG i-k ~IMG} i, by combining may generate a high-resolution image data Himg _i .

The high resolution image generation unit 17, the image data IMG _i supplied from the imaging device 30, by combining the processed image data IMG' _i supplied from the image converting unit 14, the high-resolution image data HIMG _i may be generated. In this case, the image selection unit 15 and the image memory 16 are omitted.

<Figure 10: Display (S1007)> display device 40, constituted by a frame _F 0 to F _n, and a plurality of high-resolution image data HIMG ₀ ~HIMG _n composed of a plurality of processed image data IMG' ₀ ~IMG' _n By reproducing the high-resolution frames HF _{0 to} HF _n , a normal-resolution video and a high-resolution video are displayed. When the display (S1007) ends, the remote visual inspection process in FIG. 10 ends.

According to the third embodiment, the resolution image generation unit 17 generates the high resolution image data HIMG _i . Thereby, the inspector can perform a visual inspection while comparing a normal resolution video and a high resolution video. As a result, the accuracy of visual inspection is improved.

(Fourth embodiment)
A fourth embodiment will be described. The fourth embodiment is an example of an image processing apparatus in which a high-resolution image conversion unit that performs geometric conversion on high-resolution image data is added as compared with the image processing apparatus according to the third embodiment. In addition, the description similar to said embodiment is abbreviate | omitted.

  An image processing apparatus 10 according to the fourth embodiment will be described. FIG. 12 is a block diagram illustrating a configuration of the image processing apparatus 10 according to the fourth embodiment.

  As shown in FIG. 12, the image processing apparatus 10 according to the fourth embodiment includes a movement amount calculation unit 11, a movement amount memory 12, a position estimation unit 13, an image conversion unit 14, an image selection unit 15, An image memory 16, a high resolution image generation unit 17, and a high resolution image conversion unit 18 are provided. The movement amount calculation unit 11, the movement amount memory 12, the image conversion unit 14, the image selection unit 15, and the image memory 16 are the same as those in the third embodiment.

The position estimation unit 13 is a module that estimates a suitable position. The position estimation unit 13 estimates a suitable position based on the movement amount history stored in the movement amount memory 12. Next, the position estimation unit 13 generates suitable position information PP _i representing a suitable position. Next, the position estimation unit 13 supplies the suitable position information PP _i to the image conversion unit 14 and the high resolution image conversion unit 18.

The high resolution image generation unit 17 is a module that generates high resolution image data HIMG _i . The high resolution image generation unit 17 receives the processing image data IMG ′ _i from the image conversion unit 14 and also receives setting information CONF (algorithm) from the interface 50. Then, the high resolution image generation unit 17 reads at least one target image data IMG _i from the image memory 16, using the setting information CONF (algorithm), combines the processed image data IMG' _i and the target image data IMG _i As a result, high-resolution image data HIMG _i is generated. By repeatedly generating the high-resolution image data HIMG _i , a plurality of high-resolution image data HIMG _{0 to} HIMG _n is generated. Next, the high resolution image generation unit 17 supplies a plurality of high resolution image data HIMG _{0 to} HIMG _n to the high resolution image conversion unit 18.

The high resolution image conversion unit 18 is a module that performs geometric conversion on the high resolution image data HIMG _i . The high-resolution image conversion unit 18 receives the high-resolution image data HIMG _i from the high-resolution image generation unit 17, receives the suitable position information PP _i from the position estimation unit 13, and receives the setting information CONF (algorithm) from the interface 50. Next, the high resolution image conversion unit 18 performs geometric transformation on the high resolution image data HIMG _i based on the suitable position information PP _i using the setting information CONF (algorithm), thereby corresponding to the suitable position information PP _i . High resolution processed image data HIMG ′ _i is generated. By repeatedly generating the high-resolution processed image data HIMG ′ _i , a plurality of high-resolution processed image data HIMG ′ _{0 to} HIMG ′ _n is generated. Next, the high resolution image conversion unit 18 supplies a plurality of pieces of high resolution processed image data HIMG ′ _{0 to} HIMG ′ _n to the display device 40.

The display device 40 receives a plurality of processed image data IMG ′ _{0 to} IMG ′ _n and a plurality of high resolution processed image data HIMG ′ _{0 to} HIMG ′ _n from the image processing device 10. Next, the display device 40 reproduces the frames F _{0 to} F _n configured by the plurality of processed image data IMG ′ _{0 to} IMG ′ _n to display a normal resolution video, and a plurality of high resolution processed images. A high-resolution video is displayed by reproducing high-resolution frames HF ′ _{0 to} HF ′ _n composed of data HIMG ′ _{0 to} HIMG ′ _n . As a result, a normal resolution video and a high resolution video are simultaneously displayed on the display device 40.

  An operation of the remote visual inspection apparatus 1 according to the fourth embodiment will be described. FIG. 13 is a flowchart illustrating a procedure of remote visual inspection processing according to the fourth embodiment.

  The remote visual inspection process in FIG. 13 starts when the inspector gives an instruction to the operator to operate the remote control device 20.

  <FIG. 13: Remote Operation (S1301) to Travel Calculation (S1303)> The same as in the third embodiment.

<FIG. 13: Preferential Position Estimation (S <b>1304)> The position estimation unit 13 estimates a preferable position based on the movement amount history stored in the movement amount memory 12. Next, the position estimation unit 13 generates suitable position information PP _i representing a suitable position. Next, the position estimation unit 13 supplies the suitable position information PP _i to the image conversion unit 14 and the high resolution image conversion unit 18.

  <FIG. 13: Image Selection (S1311) and Image Conversion (S1305)> The same as in the third embodiment.

<FIG. 13: High-Resolution Image Generation (S1306)> The high-resolution image generation unit 17 reads at least one target image data IMG _i from the image memory 16, and uses the setting information CONF (algorithm) to process the processed image data IMG ′. High resolution image data HIMG _i is generated by combining _i and target image data IMG _i . By repeatedly generating the high-resolution image data HIMG _i , a plurality of high-resolution image data HIMG _{0 to} HIMG _n is generated. Next, the high resolution image generation unit 17 supplies a plurality of high resolution image data HIMG _{0 to} HIMG _n to the high resolution image conversion unit 18.

<FIG. 13: High-Resolution Image Conversion (S1307)> The high-resolution image conversion unit 18 performs geometric conversion on the high-resolution image data HIMG _i based on the preferred position information PP _i using the setting information CONF (algorithm). Accordingly, to generate a high-resolution processed image data HIMG' _i corresponding to the preferred position information PP _i. By repeatedly generating the high-resolution processed image data HIMG ′ _i , a plurality of high-resolution processed image data HIMG ′ _{0 to} HIMG ′ _n is generated. For example, the geometric transformation is an affine transformation. Next, the high resolution image conversion unit 18 supplies a plurality of pieces of high resolution processed image data HIMG ′ _{0 to} HIMG ′ _n to the display device 40.

<Figure 13: Display (S1308)> display device 40, a plurality of processed image data IMG' ₀ ~IMG' frame _F composed of _n 0 to F _n, and a plurality of high-resolution processed image data HIMG' ₀ ~HIMG' By reproducing the high resolution frames HF ′ _{0 to} HF ′ _n composed of _n , a normal resolution video and a high resolution video are displayed. When the display (S1308) ends, the remote visual inspection process of FIG. 13 ends.

According to the fourth embodiment, the resolution image conversion unit 18 performs geometric conversion on the high resolution image data HIMG _i based on the suitable position information PP _i . As a result, it is possible to obtain an image taken from a viewpoint necessary for the visual inspection and having a high image quality with little camera shake. Therefore, the inspector can perform a visual inspection without feeling finer and more fatigued. As a result, the accuracy of visual inspection is improved.

(Fifth embodiment)
A fifth embodiment will be described. The fifth embodiment is an example of an image processing apparatus to which a position memory for storing a history of suitable positions is added as compared with the image processing apparatus according to the fourth embodiment. In addition, the description similar to said embodiment is abbreviate | omitted.

  An image processing apparatus 10 according to the fifth embodiment will be described. FIG. 14 is a block diagram illustrating a configuration of the image processing apparatus 10 according to the fifth embodiment.

  As illustrated in FIG. 14, the image processing apparatus 10 according to the fifth embodiment includes a movement amount calculation unit 11, a movement amount memory 12, a position estimation unit 13, an image conversion unit 14, an image selection unit 15, An image memory 16, a high resolution image generation unit 17, a high resolution image conversion unit 18, and a position memory 19 are provided. The movement amount calculation unit 11, the movement amount memory 12, the image conversion unit 14, the image selection unit 15, the image memory 16, and the high resolution image generation unit 17 are the same as those in the fourth embodiment.

The position estimation unit 13 is a module that estimates a suitable position. The position estimation unit 13 estimates a suitable position based on the movement amount history stored in the movement amount memory 12. Next, the position estimation unit 13 generates suitable position information PP _i representing a suitable position. Next, the position estimation unit 13 supplies the suitable position information PP _i to the image conversion unit 14 and stores it in the position memory 19.

The position memory 19 is a module for storing the preferred position information PP _i . That is, the position memory 19 stores a history of suitable positions.

The high resolution image conversion unit 18 is a module that performs geometric conversion on the high resolution image data HIMG _i . The high resolution image conversion unit 18 receives the high resolution image data HIMG _i from the high resolution image generation unit 17 and also receives setting information CONF (algorithm) from the interface 50. Next, the high-resolution image conversion unit 18 reads the suitable position information PP _i from the position memory 19, and uses the setting information CONF (algorithm) to perform geometric conversion on the high-resolution image data HIMG _i based on the suitable position information PP _i. by applying, to generate a high-resolution processed image data HIMG' _i corresponding to the preferred position information PP _i. By repeatedly generating the high-resolution processed image data HIMG ′ _i , a plurality of high-resolution processed image data HIMG ′ _{0 to} HIMG ′ _n is generated. Next, the high resolution image conversion unit 18 supplies a plurality of pieces of high resolution processed image data HIMG ′ _{0 to} HIMG ′ _n to the display device 40.

  The operation of the remote visual inspection apparatus 1 according to the fifth embodiment will be described. The remote visual inspection process according to the fifth embodiment is executed according to the procedure of FIG.

  The remote visual inspection process in FIG. 13 starts when the inspector gives an instruction to the operator to operate the remote control device 20.

  <FIG. 13: Remote Operation (S1301) to Travel Calculation (S1303)> The same as in the fourth embodiment.

<FIG. 13: Preferential Position Estimation (S <b>1304)> The position estimation unit 13 estimates a preferable position based on the movement amount history stored in the movement amount memory 12. Next, the position estimation unit 13 generates suitable position information PP _i representing a suitable position. Next, the position estimation unit 13 supplies the suitable position information PP _i to the image conversion unit 14 and stores it in the position memory 19.

  <FIG. 13: Image Selection (S1311), Image Conversion (S1305), High-Resolution Image Generation (S1306)> The same as in the fourth embodiment.

<FIG. 13: High-Resolution Image Conversion (S1307)> The high-resolution image conversion unit 18 reads the suitable position information PP _i from the position memory 19 and uses the setting information CONF (algorithm) based on the suitable position information PP _i. by performing geometric conversion to a high resolution image data Himg _i, to generate a high-resolution processed image data HIMG' _i corresponding to the preferred position information PP _i. By repeatedly generating the high-resolution processed image data HIMG ′ _i , a plurality of high-resolution processed image data HIMG ′ _{0 to} HIMG ′ _n is generated. Next, the high resolution image conversion unit 18 supplies a plurality of pieces of high resolution processed image data HIMG ′ _{0 to} HIMG ′ _n to the display device 40.

  <FIG. 13: Display (S1308)> The same as in the fourth embodiment. When the display (S1308) ends, the remote visual inspection process of FIG. 13 ends.

According to the fifth embodiment, the high-resolution image conversion unit 18 performs geometric conversion on the high-resolution image data HIMG _i based on the history of suitable positions. As a result, when the normal resolution video and the high resolution video are simultaneously displayed, the high resolution video frames HF _{0 to} HF with respect to the timing at which the normal resolution video frames F _{0 to} F _n are reproduced. a phenomenon in which video is intermittently caused by a delay in timing at which _n is reproduced (hereinafter referred to as “frame delay”) or frame loss of frames HF _{0 to} HF _n of a high-resolution video (hereinafter referred to as “frame advance”). ") Can be avoided. As a result, the efficiency of visual inspection is improved.

(Sixth embodiment)
A sixth embodiment will be described. The sixth embodiment is an example of an image processing device in which a prediction unit that predicts the position of the imaging device is added as compared with the image processing device according to the fifth embodiment. In addition, the description similar to said embodiment is abbreviate | omitted.

  An image processing apparatus 10 according to the sixth embodiment will be described. FIG. 15 is a block diagram illustrating a configuration of the image processing apparatus 10 according to the sixth embodiment.

  As illustrated in FIG. 15, the image processing apparatus 10 according to the sixth embodiment includes a movement amount calculation unit 11, a movement amount memory 12, a position estimation unit 13, an image conversion unit 14, an image selection unit 15, An image memory 16, a high-resolution image generation unit 17, a high-resolution image conversion unit 18, a position memory 19, and a prediction unit 21 are provided. The movement amount calculation unit 11, the movement amount memory 12, the position estimation unit 13, the image conversion unit 14, the image selection unit 15, the image memory 16, the high resolution image generation unit 17, and the position memory 19 are the same as those in the fifth embodiment. is there.

The prediction unit 21 is a module that predicts the moving direction of the imaging device 30. Based on the movement amount history stored in the movement amount memory 12, the prediction unit 21 predicts to which position and in what posture the imaging apparatus 30 moves (hereinafter referred to as “movement pattern”). The prediction information PRE _i representing the prediction result is generated. Next, the prediction unit 21 supplies the prediction information PRE _i to the high resolution image conversion unit 18 and the interface 50.

As shown in FIG. 16, the prediction information PRE _i is generated for each target image data IMG _i . Prediction information PRE _i is the prediction position information PREL _i representing the position of the imaging device 30 when generating the target image data _{IMG i + 1} is predicted from the movement pattern of the image pickup device 30 when generating the target image data IMG _i, including the prediction orientation information PREP _i representing the orientation of the imaging device 30 when generating the target image data IMG i + ₁ is predicted from the movement pattern of the image pickup device 30 when generating the target image data IMG _i.

The high resolution image conversion unit 18 is a module that performs geometric conversion on the high resolution image data HIMG _i . The high resolution image conversion unit 18 receives high resolution image data HIMG _i from the high resolution image generation unit 17, receives prediction information PRE _i from the prediction unit 21, and receives setting information CONF (algorithm) from the interface 50. Next, the high-resolution image conversion unit 18 reads the suitable position information PP _i stored in the position memory 19, and uses the setting information CONF (algorithm) to determine the high resolution based on the suitable position information PP _i and the prediction information PRE _i. by performing geometric conversion on the image data Himg _i, to generate a high-resolution processed image data HIMG' _i corresponding to the preferred position information PP _i and prediction information PRE _i. By repeatedly generating the high-resolution processed image data HIMG ′ _i , a plurality of high-resolution processed image data HIMG ′ _{0 to} HIMG ′ _n is generated. Next, the high resolution image conversion unit 18 supplies a plurality of pieces of high resolution processed image data HIMG ′ _{0 to} HIMG ′ _n to the display device 40.

The interface 50 receives the prediction information PRE _i from the prediction unit 21. Next, the interface 50 transmits the prediction information PRE _i as processing data D to the remote control device 20 or the network terminal.

  The operation of the remote visual inspection apparatus 1 according to the sixth embodiment will be described. FIG. 17 is a flowchart illustrating a procedure of a prediction process according to the sixth embodiment.

  The remote visual inspection process in FIG. 17 starts when the inspector gives an instruction to the operator to operate the remote control device 20.

  <FIG. 17: Remote Operation (S1701) to High-Resolution Image Generation (S1706) and Image Selection (S1711)> The same as in the fifth embodiment.

<FIG. 17: Prediction (S <b>1707)> The prediction unit 21 predicts the movement pattern of the imaging device 30 based on the movement amount history stored in the movement amount memory 12, thereby predicting prediction information PRE _i representing the prediction result. Is generated. Next, the prediction unit 21 supplies the prediction information PRE _i to the high-resolution image conversion unit 18 and the interface 50 at the same cycle as the sampling cycle of the imaging device 30.

For example, the prediction unit 21 approximates the motion of the imaging device 30 to a constant acceleration motion or a constant speed motion, and when the approximate movement pattern can be assumed by a Markov model, statistically and linearly captures using the Kalman filter. The movement pattern of the device 30 is predicted. Thereby, the prediction information PRE _i is generated.

<FIG. 17: High Resolution Image Conversion (S1708)> The high resolution image conversion unit 18 uses the setting information CONF (algorithm) to convert the high resolution image data HIMG _i into the high resolution image data HIMG _i based on the suitable position information PP _i and the prediction information PRE _i. by performing geometric transformation, to generate a high-resolution processed image data HIMG' _i corresponding to the preferred position information PP _i and prediction information PRE _i. By repeatedly generating the high-resolution processed image data HIMG ′ _i , a plurality of high-resolution processed image data HIMG ′ _{0 to} HIMG ′ _n is generated. Next, the high resolution image conversion unit 18 supplies a plurality of pieces of high resolution processed image data HIMG ′ _{0 to} HIMG ′ _n to the display device 40.

  <FIG. 17: Display (S1709)> The same as in the fifth embodiment.

<FIG. 17: Transmission (S <b>1710)> The interface 50 transmits the prediction information PRE _i as the processing data D to the remote control device 20 or the network terminal. On the remote control device 20 or the network terminal, a graphical model of the movement pattern of the imaging device 30 corresponding to the prediction information PRE _i and PRE _i is displayed. When the transmission (S1710) is finished, the remote visual inspection process in FIG. 17 is finished.

According to the sixth embodiment, the prediction unit 21 predicts the movement pattern of the imaging device 30 to supply the prediction information PRE _i to the high resolution image conversion unit 18. As a result, compared to the fifth embodiment, it is possible to further avoid frame delay and frame advance when displaying a normal resolution video and a high resolution video simultaneously. As a result, the efficiency of visual inspection is further improved as compared with the fifth embodiment.

Further, according to the sixth embodiment, the prediction information PRE _i is transmitted to the remote control device 20 or the network terminal via the interface 50. Thereby, the worker can operate the remote control device 20 while looking at the prediction information PRE _i . As a result, the burden on the operator's operation of the remote control device 20 is reduced.

  At least a part of the remote visual inspection apparatus 1 according to the embodiment of the present invention may be configured by hardware or software. When configured by software, a program for realizing at least a part of the functions of the remote visual inspection apparatus 1 may be stored in a recording medium such as a flexible disk or a CD-ROM and read and executed by a computer. The recording medium is not limited to a removable medium such as a magnetic disk or an optical disk, but may be a fixed recording medium such as a hard disk device or a memory.

  Moreover, you may distribute the program which implement | achieves at least one part function of the remote visual inspection apparatus 1 which concerns on embodiment of this invention via communication lines (including wireless communication), such as the internet. Further, the program may be distributed in a state where the program is encrypted, modulated or compressed, and stored in a recording medium via a wired line such as the Internet or a wireless line.

In addition, this invention is not limited to embodiment mentioned above, It deform | transforms and implements a component in the range which does not deviate from the summary. Various inventions can be formed by appropriately combining a plurality of constituent elements disclosed in the above-described embodiments. For example, you may delete a some component from all the components shown by embodiment mentioned above. Furthermore, constituent elements over different embodiments may be appropriately combined.

DESCRIPTION OF SYMBOLS 1 Remote visual inspection apparatus 10 Image processing apparatus 11 Movement amount calculation part 12 Movement amount memory 13 Position estimation part 14 Image conversion part 15 Image selection part 16 Image memory 17 High resolution image generation part 18 High resolution image conversion part 19 Position memory 20 Operation Device 21 Prediction Unit 30 Imaging Device 40 Display Device 50 Interface 60 Network

Claims (7)

An imaging unit that captures an image of the surface of the inspection object while moving;
A movement amount calculation unit that calculates a movement amount that is a change amount of the position and orientation of the imaging unit based on a difference between feature amounts of at least two image data constituting the video;
A movement amount memory for storing the movement amount history;
Based on the history of the movement amount, a position estimation unit that estimates a suitable position of the imaging unit and a posture suitable for visual inspection for an inspector;
An image conversion unit that generates processed image data corresponding to the preferred position by performing geometric transformation on the image data based on the preferred position;
A remote visual inspection comprising: a high-resolution image generation unit that generates high-resolution image data having a higher resolution than the resolution of the processed image data by combining the image data and the processed image data apparatus. An image selection unit that determines whether the image data meets a predetermined selection condition based on the movement amount;
An image memory for storing image data that conforms to the selection conditions,
The remote visual inspection apparatus according to claim 1, wherein the image conversion unit performs geometric conversion on image data stored in the image memory. By performing geometric conversion on the high-resolution image data based on the preferred position, further comprising a high resolution image converter for generating a high-resolution processed image data corresponding to the preferred position, according to claim 1 or 2 remotely according Visual inspection device. A movement amount memory for storing a history of the preferred position;
The remote visual inspection apparatus according to claim 3 , wherein the high-resolution image conversion unit performs geometric conversion on the high-resolution image data based on the movement amount history. A prediction unit that predicts a movement pattern of the imaging unit based on the movement amount history;
The remote visual inspection apparatus according to claim 4 , wherein the high-resolution image conversion unit performs geometric conversion on the high-resolution image data based on the movement pattern. The remote visual inspection apparatus according to claim 5 , wherein the prediction unit transmits the movement pattern to a network terminal connected via a network. By playing frame constituted by the processed image data generated by the image conversion unit further comprises a display unit for reproducing an image of said object surface, according to any one of claims 1 to 6 Remote visual inspection equipment.

Images