JP4319333B2 - Compressed image data restoration device, compressed image data restoration method, and medium recorded with compressed image data restoration program - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a compressed image data restoration device, a compressed image data restoration method, and a medium on which a compressed image data restoration program is recorded.
[0002]
[Prior art]
In recent years, when a photograph is processed on a computer, compression processing is used to reduce the file capacity. As an example, what is called a JPEG image is encoded by replacing an 8 × 8 pixel area as a unit with a predetermined pattern.
[0003]
On the other hand, such compressed image data is often stored in various recording media such as a hard disk, but cannot be corrected by a file error correction function such as an operating system at the time of reading / writing and storing in the recording medium. The number of errors is increasing.
It is rare for an error to occur several consecutive bytes per image, and usually only one byte occurs independently. If the image data is also uncompressed, it is only necessary to reproduce the one pixel in which an error has occurred. However, if the image data is compressed, the pixel after the error has an effect. Also, since images are sequentially compressed from the top according to the image status, it is extremely difficult to identify data in which an error has occurred. Furthermore, even if the error location is searched from the beginning of the file, there is no means to automatically determine whether or not the error has been repaired, so it is necessary to work manually, which takes a very long time. .
[0004]
[Problems to be solved by the invention]
As described above, conventionally, even if an error occurs in the compressed image data, the error location cannot be specified, so correction must be attempted from the beginning position of the file. However, today, when files are getting larger, it takes a very long time, and there is a problem that it is practically impossible.
The present invention has been made in view of the above problems, and provides a compressed image data restoration device, a compressed image data restoration method, and a medium recorded with a compressed image data restoration program capable of restoring compressed image data in a short time. With the goal.
[0005]
[Means for Solving the Problems]
In order to achieve the above object, the invention according to claim 1 is directed to an instruction operation means for restoring and displaying compressed image data and receiving an instruction operation for an image breakage point, and for the restored image data of the designated image breakage point. Initial setting means for setting the estimated damage point in the compressed image data corresponding to the relative position information, and when the compressed image data of the damaged estimated point is changed and restored based on the changed compressed image data A first breakage point verification means for detecting the difference between the case where the restoration process is performed based on the original compressed image data and verifying the damage estimated point, and changing the compressed image data of the damage estimated point The estimated damage point is determined based on the displacement of the image at the boundary between the damaged image region and the undamaged image region based on the image damaged point when the restoration processing is performed based on the later compressed image data. The second breakage point verification means, the first breakage point verification means, and the second breakage point verification means to be verified, and the breakage point approaching the original breakage point based on the verification result And a point resetting means.
[0006]
In the invention according to claim 1 configured as described above, when the instruction operation unit restores and displays the compressed image data, the operator instructs the image breakage point, and when the instruction operation unit accepts the instruction operation, The initial setting means sets a damage estimation point in the compressed image data corresponding to the relative position information in the restored image data of the specified image damage point. Here, the first breakage point verification means changes the compressed image data of the breakage estimation point, and when restored based on the compressed image data after the change and when restored based on the original compressed image data The second damage point verification means changes the compressed image data of the damage estimation point and restores based on the compressed image data after the change. The estimated damage point is verified based on the displacement state of the image at the boundary between the damaged image area and the undamaged image area based on the image damaged point. Therefore, the breakage point resetting unit appropriately controls the first breakage point verification unit and the second breakage point verification unit, and brings the breakage estimation point closer to the original breakage point based on the verification result.
[0007]
For example, when the first breakage point verification means is used, the difference between the case where the restoration process is performed based on the compressed image data after the change and the case where the restoration process is performed based on the original compressed image data is detected. It is possible to roughly grasp whether the estimated point is on the front side or the rear side of the original breakage point.
However, even if the first breakage point verification means is used, even if the difference between the breakage estimation point and the original breakage point can be found, correct data cannot be verified.
[0008]
On the other hand, when the second breakage point verification means is used, at the boundary between the damaged image area and the undamaged image area based on the image breakage point when the restoration process is performed based on the compressed image data after the change. In order to detect the displacement state of the image, it can be seen that the estimated damage point and the change data when the displacement is small are the original correct data.
However, since the second breakage point verification means has a large work load, it is not efficient to verify from the beginning of the file without any address.
[0009]
As a result of considering both of these characteristics, the invention according to claim 2 is the compressed image data restoration device according to claim 1, wherein the breakage point resetting means includes the breakage estimation point and the original breakage point. When the distance is far, the first breakage point verification means is implemented, and when the distance is close, the second breakage point verification means is implemented.
In the invention according to claim 2 configured as described above, the estimated damage point can be obtained with a light load by performing the first damaged point verification means when the estimated damage point is far from the original damaged point. After approaching and close to the original breakage point, the second breakage point verification means locates the original breakage point and correct data.
[0010]
By the way, since the compressed image data is compressed from the head in accordance with the state of the original image, it is not unambiguous how the influence result when the data is changed is reflected. That is, when the data near the beginning of the file is changed, the portion near the beginning of the restored image data does not always change, so when the data at a certain damage estimation point is changed, it is behind the original damage point. Just because the data has changed, we cannot say that the damage estimation point is ahead.
[0011]
Therefore, according to a third aspect of the present invention, in the compressed image data restoration device according to the first or second aspect, the first breakage point verification means includes a plurality of breakage estimation points as a reference. The compressed image data at the point is changed, the difference between the case where the restoration process is performed based on the compressed image data after the change and the case where the restoration process is performed based on the original compressed image data is detected, and the damage estimation point is verified. As a configuration.
In the invention according to claim 3 configured as described above, the compressed image data at a plurality of points based on the estimated damage point is changed. For example, the compressed image data is changed for several pixels on the back side of the file with reference to the estimated damage point, and the restoration process is performed based on the changed compressed image data. In this case, in general, a change in a later pixel tends to produce a difference on the rear side compared to a case where restoration processing is performed based on the original compressed image data, but it is possible to reverse some pixels. There is also sex. Even if such a discrepancy occurs, if the result of several pixels is comprehensively considered, it can be correctly determined whether the damage estimation point is ahead or behind the original damage point.
[0012]
The second breakage point verification means detects the image displacement state at the boundary between the damaged image area and the non-damaged image area. Although it can be accurately determined if detailed verification is made at the entire boundary, the load is extremely high. Become bigger. On the other hand, if the judgment is made with only a few boundary regions, the processing speed is improved, but the verification accuracy is lowered.
Therefore, the invention according to claim 4 is the compressed image data restoration device according to any one of claims 1 to 3, wherein the second breakage point verification means can designate a verification mode, Corresponding to the verification mode, the detection density at the boundary between the damaged image region and the non-damaged image region is changed.
[0013]
In the invention according to claim 4 configured as described above, it is possible to change the detection density at the boundary between the damaged image region and the undamaged image region by selecting the verification mode. You can plan.
Also, due to the nature of the compressed image data, it is not always possible to restore the compressed image when the data is changed at an arbitrary point. If the data cannot be restored, the combination of the damaged point and the changed data is not correct. I can say that. Therefore, according to a fifth aspect of the present invention, in the compressed image data restoration device according to any one of the first to fourth aspects, the breakage point resetting means includes the first breakage point verification means and the first breakage point verification means. When the compressed image data of the estimated damage point is changed by the second damaged point verification means, if the restoration is impossible based on the compressed image data after the change, the damaged point is reset.
[0014]
In the invention according to claim 5 configured as described above, even if the compressed image data of the estimated damage point is changed by the first damaged point verification unit and the second damaged point verification unit, If the data cannot be restored based on the compressed image data, the data should be incorrect, and the damaged point is reset without proceeding with the verification.
Further, the invention according to claim 6 is the compressed image data restoration device according to any one of claims 1 to 5, wherein the breakage point resetting means is located before the breakage estimation point and the original breakage point. The damage estimation point is reset so as to be close to each other.
[0015]
In the invention according to claim 6 configured as described above, when the estimated damage point is brought close to the original damaged point, the estimated damage point is reset from the front side of the file. .
Although the compressed image data can be applied to various types, the invention according to claim 7 is the compressed image data restoration device according to any one of claims 1 to 6, wherein the compressed image data is a still image. The invention according to claim 8 is directed to a compressed moving image as the compressed image data in the compressed image data restoration device according to any one of claims 1 to 6.
[0016]
On the other hand, it is possible not only to repair the damage by the above-described method, but also to be able to repair other damage patterns. As an example, the invention according to claim 9 includes the above-mentioned claim. The compressed image data restoration device according to any one of claims 1 to 8, wherein the instruction operation means replaces predetermined data and displays an image if display is impossible when displaying damaged compressed image data. It is configured so that it can be performed.
[0017]
In the invention according to claim 9 configured as described above, if display is impossible when displaying damaged compressed image data, the instruction operation means replaces predetermined data so that an image can be displayed. For example, the damage in a fixed part such as a header part of a file may be displayed by discriminating the type of image with another means and replacing it with standard data. In addition, if the damage is not limited to the header portion but in a known pattern, it is possible to examine only a certain range of data and try to replace the data.
[0018]
In addition, although it is easy and reliable for a human to perform an instruction operation for an image breakage point, it is not impossible to perform automation. As an example, in the invention according to claim 10, in the compressed image data restoration device according to any one of claims 1 to 9, the instruction operation means restores and displays the compressed image data. At this time, the presence or absence of an image breakage point is detected, and when it is detected, the instruction operation is accepted.
[0019]
In the invention according to claim 10 configured as described above, the instruction operation means restores the compressed image data before receiving the instruction operation. At this time, the presence or absence of an image breakage point is detected. In some cases, a typical pattern of damage can be detected relatively easily in compressed image data. Transition to processing.
As described above, it is not always necessary that the method for searching for the estimation of the breakage point and the elucidation of the original data of the compressed image data is limited to a substantial device, and it can be easily understood that the method also functions as a method. For this reason, the invention according to claim 11 restores and displays the compressed image data, and receives an instruction operation step for receiving an instruction operation for an image breakage point, and a relative position of the instructed image breakage point in the restored image data An initial setting step for setting a damage estimation point in the compressed image data corresponding to the information, and a case where the compressed image data of the damage estimation point is changed and restored based on the compressed image data after the change and the original compression A first breakage point verification step for detecting a difference from the case of restoration processing based on image data and verifying the breakage estimation point; and changing the compressed image data of the breakage estimation point and changing the compressed image after the change Secondly, the damage estimation point is verified based on a displacement state of the image at the boundary between the damaged image area and the undamaged image area based on the image damage point when the restoration process is performed based on the data. Control of execution of the failure point verification step, the first failure point verification step, and the second failure point verification step, and based on the verification results, the failure point re-establishment approaches the original failure point. And a setting step.
[0020]
That is, it is not necessarily limited to a substantial apparatus, and there is no difference that the method is also effective.
By the way, such a compressed image data restoration device may exist alone, or may be used in a state of being incorporated in a certain device. Is included. Therefore, it can be changed as appropriate, such as software or hardware.
[0021]
When the software of the compressed image data restoration apparatus is implemented as an embodiment of the idea of the invention, it naturally exists on a recording medium on which such software is recorded and must be used.
As an example, the invention according to claim 12 is a medium on which a compressed image data restoration program for restoring compressed image data by a computer is recorded, the compressed image data is restored and displayed, and an image breakage point instruction operation is performed. An instruction operation function for accepting an image, an initial setting function for setting a damage estimation point in the compressed image data corresponding to the relative position information in the restored image data of the specified image damage point, and compression of the damage estimation point First, the image data is changed to detect the difference between the case where the restoration processing is performed based on the compressed image data after the change and the case where the restoration processing is performed based on the original compression image data, and the above damage estimation point is verified. Based on the damaged point verification function and the image damaged point when the compressed image data of the estimated damage point is changed and restored based on the changed compressed image data. A second damage point verification function for verifying the damage estimation point based on a displacement state of an image at a boundary between the damaged image area and the undamaged image area, the first damage point verification function, and the second And a break point resetting function for controlling the execution of the breakage point verification function and bringing the estimated breakage point closer to the original breakage point based on the verification result.
[0022]
Of course, the recording medium may be a magnetic recording medium, a magneto-optical recording medium, or any recording medium that will be developed in the future. In addition, the duplication stages such as the primary duplication product and the secondary duplication product are equivalent without any question. In addition, even when the communication method is used as a supply method, the present invention is not changed.
Further, even when a part is software and a part is realized by hardware, the idea of the invention is not completely different, and a part is stored on a recording medium and is appropriately changed as necessary. It may be in the form of being read.
[0023]
When the present invention is realized by software, a configuration using hardware or an operating system may be used, or may be realized separately from these. For example, even if it is a process of inputting image data to perform an interpolation process, if the implementation method can call and process a predetermined function in the operating system, the hardware can be called without calling such a function. It is also possible to input from. It can be understood that the present invention can be implemented only by this program in the process in which the program is recorded on the medium and distributed even though it is actually realized under the intervention of the operating system.
[0024]
When the present invention is implemented by software, the present invention is not only realized as a medium storing a program, but the present invention is naturally realized as a program itself, and the program itself is also included in the present invention.
[0025]
【The invention's effect】
As described above, the present invention can provide a compressed image data restoration device that can repair damaged compressed image data, which has been impossible in the past, in a short time.
[0026]
Further, according to the invention of claim 2, the first breakage point verification means and the second breakage point verification means can be used efficiently and repaired in a short time.
Furthermore, according to the third aspect of the present invention, it is possible to accurately verify the estimated damage point in a short time without being influenced by chance.
Furthermore, according to the invention concerning Claim 4, a precision and processing time can be harmonized according to the objective.
Furthermore, according to the fifth aspect of the present invention, it is possible to quickly give up to the estimated damage points that cannot be restored and to shorten the processing time.
[0027]
Furthermore, according to the invention according to claim 6, by making the estimated damage point close to the original position, the previous determination result can be used, for example, when processing is performed based on a plurality of estimated damage points, Efficiency can be improved.
Furthermore, according to the seventh aspect of the present invention, it is possible to restore a still image.
Furthermore, according to the invention concerning Claim 8, it becomes possible to restore | restore a moving image.
Furthermore, according to the invention of claim 9, there is a possibility that a compressed image that cannot be displayed can be repaired.
[0028]
Furthermore, according to the tenth aspect of the present invention, there is a possibility that an instruction operation for an image breakage point by a human can be omitted.
Furthermore, the invention according to claim 11 can provide a compressed image data restoration method capable of producing the same effect. According to the invention according to claim 12, a medium on which a compressed image data restoration program is recorded can be provided. Can be provided.
[0029]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings.
FIG. 1 is a diagram corresponding to a claim showing a compressed image data restoration apparatus of the present invention.
Digital photograph images taken with a digital still camera or the like are stored as compressed image data files such as JPEG. This compressed image data restoration device restores when a part of such a compressed image data file is rewritten to unintended data, and the instruction operation means C1 performs an instruction operation of a damaged point in the restored image. Receiving and initial setting means C2 calculates the damage estimated point in the compressed image data from the relative position of the instructed damaged point on the restored image, and the first damaged point verifying means C3 and the second that verify the estimated damage point by different methods. The breakage point resetting means C5 appropriately clears the breakage point verification means C4 to find the original breakage point in the compressed image data.
[0030]
In the present embodiment, a computer system 10 is employed as an example of hardware that realizes such a compressed image data restoration apparatus.
FIG. 2 is a block diagram showing the computer system 10.
The computer system 10 includes a scanner 11a, a digital still camera 11b, and a video camera 11c as image input devices, and is connected to a computer main body 12. Each input device can generate image data in which an image is expressed by a dot matrix pixel and output the image data to the computer main body 12. Here, the image data is displayed in 256 gradations for each of the three primary colors of RGB. About 16.7 million colors can be expressed.
[0031]
A floppy disk drive 13a, a hard disk 13b, and a CD-ROM drive 13c as external auxiliary storage devices are connected to the computer main body 12, and main programs related to the system are recorded on the hard disk 13b. Necessary programs and the like can be read from a CD-ROM or the like as appropriate.
In addition, a modem 14a is connected as a communication device for connecting the computer main body 12 to an external network or the like, and it can be connected to an external network via the public communication line, and can be installed by downloading software and data. It has become. In this example, the modem 14a accesses the outside via a telephone line, but it is also possible to adopt a configuration where the network is accessed via a LAN adapter. In addition, a keyboard 15a and a mouse 15b are also connected for operating the computer main body 12.
[0032]
Furthermore, a display 17a and a color printer 17b are provided as image output devices. The display 17a has a display area of 800 pixels in the horizontal direction and 600 pixels in the vertical direction, and the above-described display of 16.7 million colors can be performed for each pixel. Of course, this resolution is only an example, and can be changed as appropriate, such as 640 × 480 pixels or 1024 × 768 pixels.
[0033]
On the other hand, in order to display or output to the image output device while inputting an image using such an image input device, a predetermined program is executed in the computer main body 12. Among them, an operating system (OS) 12a is operating as a basic program, and the operating system 12a performs a print output to a display driver (DSP DRV) 12b for displaying on the display 17a and a color printer 17b. A printer driver (PRT DRV) 12c is installed. These types of drivers 12b and 12c depend on the models of the display 17a and the color printer 17b, and can be added to and changed from the operating system 12a according to the respective models. In addition, depending on the model, additional functions beyond standard processing can be realized. That is, various additional processes within an allowable range can be realized while maintaining a common processing system on the standard system called the operating system 12a.
[0034]
The application 12d is executed on the operating system 12a as the basic program. The processing contents of the application 12d are various. The operation of the keyboard 15a and the mouse 15b as operation devices is monitored, and when operated, various external devices are appropriately controlled to execute corresponding arithmetic processing, Furthermore, the processing result is displayed on the display 17a or output to the color printer 17b.
[0035]
In such a computer system 10, compressed image data is acquired by a digital still camera 11 b or the like which is an image input device, predetermined image processing is executed by an application 12 d, and then displayed on a display 17 a or a color printer 17 b as an image output device. Is possible.
If a part of the compressed image data is damaged, it cannot be restored to the original image, and display and printing will not remain. However, even if a part of the file is damaged, a 1-byte data error is rarely generated. The compressed image data cannot be restored until a large amount of data in the file is lost and cannot be restored. However, even if the compressed image data cannot be displayed at all, if it is about 1 byte, it may be displayed by performing a total attack from the beginning of the file, so there is a possibility that it can be restored by the present invention.
[0036]
In particular, depending on the type of file corruption, the repair may be relatively simple. For example, data indicating the type of the compressed image data file may be damaged. In the example of a JPEG image, data indicating that the data file is JPEG-compressed is written at the beginning of the file, but even if one byte of the data file is damaged, it is unclear that it is JPEG-compressed. And will be excluded from restoration. However, it can be strongly inferred that the image data file is more than the operator has pointed out as a damaged compressed image, and it can often be presumed that JPEG compression is performed from the extension of the file. In such a case, the damaged data may be replaced with other data in the fixed part of the data file so that an image can be displayed, and then repair of the damaged data in the other part may be attempted. Of course, when the image can be completely displayed by repairing only the fixed portion, the repair is completed. Of course, such replacement of data is performed by the instruction operation means C1. It should be noted that even if it is not necessarily a fixed portion, it may be attempted to display or not by replacing data within a predetermined range in the data file.
[0037]
FIG. 3 is a flowchart showing the compressed image data restoration program as the application 12d.
When the compressed image data restoration program is activated, first, in step 100, a compressed image data file is designated. That is, the program is a process of monitoring the operation of the keyboard 15a and mouse 15b by the operator via the operating system 12a and waiting for selection of compressed image data. When a specified operation for specifying a predetermined compressed image data file is performed using a file open dialog or an explorer, the compressed image data file is restored to a bitmap image in step 200 and displayed on the display 17a. The restored bitmap image is stored in the hard disk 13b. The hard disk 13b is saved because the bitmap image is repeatedly referred to in later steps, and the RAM disk area or the like may be designated and saved as necessary.
[0038]
Next, in step 300, an operation for instructing the position of the damaged block is input. FIG. 4 shows a display example on the display 17a when one byte is damaged in a general JPEG compressed image data file. In the drawing, the upper half of the unbroken image area 17a1 and the lower half of the damaged image area 17a2 are shown separately. However, the upper half of the unbroken image area 17a1 is displayed normally, and the lower half of the damaged image area 17a2 is displayed. It is assumed that it is not displayed normally. FIG. 5 is an enlarged view of the vicinity of the damaged block 17a5 that has started not to be displayed normally. The JPEG image is compressed in units of block areas of every 8 × 8 pixels, and is compressed from the upper left to the right and sequentially downward, so the right side and the lower side of the damaged block 17a5 including the damaged point Area is not displayed properly.
[0039]
The operator operates the mouse 15b to point to the broken block 17a5 including the broken point with the arrow pointer 17a3, and clicks. At this time, an 8 × 8 pixel area 17a4 indicated by the arrow pointer 17a3 is displayed in reverse so that the operator can correctly indicate a damaged area on the display by the arrow pointer 17a3. Note that the software processing in step 300 for accepting this operation and related hardware constitute the instruction operation means C1. In the present embodiment, the mouse 15b is used to indicate the area. However, the area of 8 × 8 pixels is moved by the cursor, or the area display (A, B, C,. 3 ...) may be input (for example, "B-4").
[0040]
In addition to this, the instruction operation can be further automatically performed. As described above, in the case of a JPEG image, the tendency of a color suddenly changes from a certain pixel as a damaged pattern to an upper region and a lower region in the right direction in the horizontal direction. The tendency of the color suddenly changes between the upper area and the lower area in the left direction. Therefore, the presence / absence of such characteristic breakage is detected in advance, and if breakage is detected, the following processing may be executed assuming that the above-described breakage area instruction has been given. In this sense, the instruction operation means C1 can be configured to include automation of a damaged area instruction.
[0041]
Returning to FIG. 3, in step 400, an estimated damage point is calculated from the relative position information on the display image of the specified damaged area. Originally, compressed image data is gradually generated from the front of the file according to the image status, so even if the relative position information on the display image of the damaged area is obtained, the data file depends on the image status. The corresponding position above should be unknown. In particular, the tendency is further emphasized when the stream distribution technology is adopted so as to improve the appearance as much as possible even during data distribution.
[0042]
However, even if there are fluctuations, an approximate trend can often be grasped. FIG. 6 shows the relative relationship between the change positions obtained by restoring the image when a part of the data file is rewritten. In the figure, the horizontal axis is the position information that changes the data and shows the first damage estimated point in the data file as a reference, and the vertical axis changes based on the damage position corresponding to the first damage estimated point The pixel position information is shown. Actually, even if data of the same position is changed using a data file of the same size depending on the image, the change position generated on the display screen varies. However, as is apparent from this graph display, rough estimation is possible. In this embodiment, the least square approximation function is
Y = 0.003507 * X ** 2-1.425665 * X
In step 300, a damage estimation point is calculated using the same calculation formula. Note that it is impossible to obtain a complete answer with this calculation formula, and it is possible to change the calculation formula to a simpler primary formula, or to create a table in advance and use it as a simple quick reference table. . In any case, the software processing for calculating the estimated damage point from the relative position information on the display image of the damaged area designated as described above and the related hardware constitute the initial setting means C2.
[0043]
Subsequently, in step 500, a process for bringing the estimated damage point calculated as described above closer to the original damaged point is performed. FIG. 7 shows a detailed flowchart thereof. In step 502, the calculated estimated damage point is obtained through the work area and set to the estimated damage point for the work. Subsequently, in step 504, the data of the estimated damage point is changed, and in step 506, restoration of the bitmap image is attempted using the changed data file.
[0044]
However, since the data at an arbitrary position has been changed, the restoration is not always possible, and only when the restoration is possible, the bitmap image restored in step 510 and the step 200 after the judgment in step 508 are performed. Compare the image saved in step 1 to determine the difference position. That is, if the data is changed with a certain point as a damage estimation point, the bitmap image restored using the data file should be different from the one before the change. Since the damaged area on the bitmap image is determined by pointing with the mouse 15b, it can be determined whether the position of the difference is forward or backward by comparing with the position. If the different position is on the rear side, it can be said that the estimated damage point is on the rear side in the data file rather than the damaged point, as estimated from the relationship shown in FIG.
[0045]
In step 512 and step 514, the positional relationship between the different positions and the breakage point is determined, and the breakage estimation point is moved. That is, if it is behind the estimated damage point, the estimated damage point is returned further forward in step 518 after the determination in step 512. If it is nearer than the estimated damage point but is still farther than the estimated damage point, the estimated damage point is moved slightly backward in step 516 through the determination in step 512 and step 514. Normally, the processing from step 512 to step 516 is repeated so that the estimated damage point gradually approaches the damaged point from the front. When the estimated damage point finally approaches the damaged point, the determination of step 514 is followed to end the damaged point approaching process.
[0046]
In this way, whether or not the above-mentioned damage estimation point is before the damage point based on the difference position between the bitmap image restored after changing the data with a certain point as the damage estimation point and the restored bitmap image. The software processing and the related hardware for determining the first breakage point verification means C3.
In this example, the bitmap image is restored to the end based on the changed data file, but the processing can be stopped when it is found that the difference position is far from the breakage point. It is also possible to set a plurality of points on the basis of the estimated damage points and perform processing simultaneously. For example, when a certain damage estimation point is set and the bitmap image is restored, and the data of the damage estimation point is changed and restored, the damage position at which the bitmap image has changed thereby indicates the damage position indicated by the mouse 15b. If it is nearer and not closer than the above, the damage estimation point is set further forward while returning the data of the damage estimation point to the original data. At this time, the bitmap image restored until then is used as it is. In this way, it is not necessary to repeat the calculation for restoration each time the estimated damage point is changed, so that the processing can be speeded up. In this sense, the setting of the damage estimation point has an advantage in the case of always approaching from the near side. Of course, whether or not the above processing is possible depends on the nature of the compressed image data, and is not necessarily realizable in all cases.
[0047]
Further, if the estimated damage point is moved backward, the above-mentioned different position is not necessarily moved backward, so that it is often possible to obtain a correct result by comprehensively judging using a plurality of points. Therefore, it is meaningful to verify using a plurality of points based on the estimated damage point.
When the breakpoint approaching process is completed, the total attack mode is started in step 600 as shown in FIG. FIG. 8 shows this total attack mode in more detail. The damage estimation point set last in step 602 is set as a damage estimation point for a predetermined workpiece, and in step 616, the damage estimation point is sequentially set to one byte. While moving each step, the data of the estimated damage point is changed between “0” to “255” between step 604 and step 614 to determine whether or not the image has been correctly restored.
[0048]
Whether or not the image has been correctly restored is determined by changing the data of the estimated damage point in step 606 to restore the bitmap image, and then determining the smoothness of the image around the damaged area in step 608. ing. FIG. 9 shows in detail the smoothness determination process of this image, and FIG. 10 shows the boundaries used for the determination.
In this example, the boundary portion between the damaged block 17a5 and the upper half of the undamaged image area 17a1 is verified in Step 620 to Step 626, and the upper half of the undamaged image area 17a1 and the lower half of the upper half of the image are damaged. The boundary with the damaged image area 17a2 is verified.
[0049]
First, in step 620, a pointer is set at the boundary of the damaged block 17a5. Originally, smoothness may be determined for all boundary pixels, but the amount of processing increases accordingly. Accordingly, it is determined whether to perform the process for all the pixels or to perform the thinning process based on the balance between the processing speed and the accuracy. Of course, the operator may be allowed to select. In step 622, the color change amount at the pointer position is calculated. In FIG. 10, the average of the colors of two pixels in each of the undamaged image area 17a1 and the damaged image area 17a2 is calculated for the four pixels across the boundary, and the difference between the respective average values is stored as a color change amount. Accumulate. In step 624, the pointer is moved, and the process is repeated until it is determined in step 626 that all the boundaries have been completed. Referring to the figure, the boundary between the damaged block 17a5 and the upper half of the undamaged image area 17a1 is 16 pixels (= 8 pixels + 8 pixels), and when calculating for all the pixels, the color change amount at 16 pixels is integrated. Will be.
[0050]
Thereafter, in step 628, the same calculation is performed at the boundary where the upper side of the damaged block 17a5 is extended backward. However, since this boundary is expected to be long, the calculation is executed by thinning out every 10 pixels or every 20 pixels. In step 630, the same calculation is performed at the boundary where the lower side of the damaged block 17a5 is extended forward.
In step 632, smoothness is calculated using the color change amount stored and integrated at each boundary. The smoothness here is simply used as a parameter and does not require a special formula. For example, an average value obtained by dividing the integrated amount by the sum of the calculated pixels may be used, or a simple integrated amount may be used. Furthermore, regarding the color change amount itself, the change tendency of the two pixels in the upper half of the undamaged image area 17a1 and the change tendency of the two pixels in the lower half of the damaged image area 17a2 may be calculated and the difference may be taken. For example, if the upper two pixels become gradually brighter, it is often considered that the lower two pixels tend to become gradually brighter. And if the change tendency as mentioned above is taken, both will correspond in this case. Of course, the use of two pixels is arbitrary, and the number of pixels can be increased or decreased.
[0051]
In step 610, the smoothness calculated in this way is used to determine whether or not the reference value has been exceeded. In step 618, in step 618, the estimated damage point at that time and the changed data are stored and stored. Exit attack mode.
When the damage estimation point is shifted byte by byte as described above, the boundary between the undamaged image area 17a1 and the damaged image area 17a2 for all values “0” to “255” that can be taken for each byte. The software processing for determining whether or not the image touches smoothly and the related hardware constitute the second breakage point verification means C4.
[0052]
In the present embodiment, the displacement state of the image at the boundary is determined using the color change amount, but it is determined whether or not the undamaged image area 17a1 and the damaged image area 17a2 are smoothly connected. There are various methods. FIG. 11 schematically shows another method. In this example, the road 17a6 is shown in the upper half of the unbroken image area 17a1, but the road 17a6 is not displayed in the damaged image area 17a2. However, it is unlikely that the road is actually broken, and in the situation where a known object such as a road is photographed, it is possible to infer the road shape of the damaged place from the property of the object as shown in FIG. Is possible. When the data is changed at the estimated damage point, if the road is restored to a shape that fits the guess, along with the information that determines the smoothness, the criteria for image restoration can be used to expect more accurate data restoration. It is.
[0053]
In addition, the break point approaching process is performed first as the processing order, and at that time, the break point is reset based on the verification result of the first break point verification means C3 described above, Whether or not the image has been restored based on the verification result of the second breakage point verification means C4 while shifting to the total attack mode and shifting the breakage estimated point by 1 byte from when the image approaches the point before the original breakage point Although it is determined, the processing flow for setting the estimated damage point and the related hardware constitute the damaged point resetting means C5. Of course, it is possible to execute the same processing in different procedures depending on software, and the processing procedure of the present embodiment described above is merely an example.
[0054]
In the above-described embodiment, only a still image is handled, but a moving image can be considered in the same manner. FIG. 12 shows how data appears when data is damaged in the case of a moving image. Although the first frame is normally displayed, an image of meaninglessness is suddenly displayed on the second frame, and the same image is displayed on the third frame. However, it disappears suddenly in the fourth frame.
In such a case, a moving image is displayed as a frame and an abnormal frame is indicated by the mouse 15b as an image breakage point. Then, the relative position of the image breakage point in all data files is checked in frame number units to set the damage estimation point of the data file, and the data of the damage estimation point is rewritten. Based on the rewritten data file, the frame-by-frame image is restored again, and it is examined at which point the difference from the first image appears. If it is behind the instructed image damage point, the damage estimation point is moved to the front side of the data file, and if it is in front of the instructed image damage point, the damage estimation point is gradually moved to the rear side.
[0055]
Then, when approaching the frame where an abnormality has occurred, the total attack mode is set, and the displacement state of the image is determined for each frame. In the case of a moving image, the periphery of the damaged area may be checked, or the presence or absence of a sudden image change may be checked. In the natural world, there is a limit to the change of the image, so it is determined whether or not the image has changed beyond that. In addition, in the case of a moving image, data is often sent for the changed image, so the influence of the damaged data occurs over a plurality of frames at the same location. Therefore, if there is a change in the image only in all the frames designated as damaged images, there is a high possibility that it may be determined that the data has been restored.
[0056]
In this way, after instructing the damaged image, try to restore the compressed image by changing the data of the damage estimation point while setting the damage estimation point, and from the position different from the original image data, the damage estimation point Of the boundary between the damaged image area and the damaged image area when the data of the estimated damage point is changed when the estimated damage point approaches the damaged point. By examining the situation, it has become possible to automatically repair damaged compressed image data in a short time.
[Brief description of the drawings]
FIG. 1 is a claim correspondence diagram showing a compressed image data restoration device according to an embodiment of the present invention.
FIG. 2 is a block diagram of a computer system as an example of hardware for realizing a compressed image data restoration device.
FIG. 3 is a flowchart showing a procedure for restoring compressed image data.
FIG. 4 is a diagram showing a screen display when restoring damaged image data and indicating a damaged block.
FIG. 5 is an enlarged view near a damaged block.
FIG. 6 is a diagram illustrating a correlation between a change position of a data file and a change position of a restored image.
FIG. 7 is a flowchart of breakage point approach processing.
FIG. 8 is a flowchart of a total attack mode.
FIG. 9 is a flowchart of image smoothness determination processing;
FIG. 10 is an enlarged view of a boundary portion referred to in image smoothness determination processing.
FIG. 11 is a diagram illustrating another method for detecting a displacement state of an image.
FIG. 12 is a diagram illustrating an image change that occurs when the compressed image data in the case of a moving image is restored.
[Explanation of symbols]
C1 ... Instruction operation means
C2 ... Initial setting means
C3 ... Damage point verification means
C4 ... Broken point verification means
C5 ... Broken point resetting means
10. Computer system
11a ... Scanner
11b ... Digital still camera
11c ... Video camera
12 ... Computer body
12a ... Operating system
12b ... Printer driver
12c ... Display driver
12d Application
13a ... floppy disk drive
13b ... Hard disk
13c ... CD-ROM drive
14a ... modem
15a ... Keyboard
15b ... Mouse
17a ... Display
17a1 ... Undamaged image area
17a2 ... Damaged image area
17a3 ... Arrow pointer
17a4 ... area
17a5 ... Damaged block
17a6 ... Road
17b ... Color printer

Claims (12)

Instruction operation means for restoring and displaying the compressed image data and receiving an instruction operation for an image breakage point;
Initial setting means for setting a damage estimation point in the compressed image data corresponding to the relative position information in the restored image data of the indicated image damage point;
The damage estimation point is detected by detecting a difference between the case where the compressed image data of the damage estimation point is changed and the restoration process is performed based on the compressed image data after the change and the restoration process is performed based on the original compression image data. A first breakage point verification means for verifying,
Displacement of the image at the boundary between the damaged image area and the undamaged image area when the compressed image data at the estimated damage point is changed and restored based on the compressed image data after the change A second breakage point verification means for verifying the breakage estimation point based on the situation;
A failure point resetting unit for controlling the first breakage point verification unit and the second breakage point verification unit and bringing the breakage estimation point closer to the original breakage point based on the verification result; A compressed image data restoration device characterized.   The compressed image data restoration device according to claim 1, wherein the breakage point resetting unit performs the first breakage point verification unit when the breakage estimation point and the original breakage point are far from each other. A compressed image data restoration device characterized in that the second breakage point verification means is implemented.   3. The compressed image data restoration device according to claim 1, wherein the first breakage point verification unit changes the compressed image data at a plurality of points based on the breakage estimated point, and after the change A compressed image data restoration device that detects a difference between a case where restoration processing is performed based on the compressed image data and a case where restoration processing is performed based on the original compression image data, and verifies the estimated damage point.   The compressed image data restoration device according to any one of claims 1 to 3, wherein the second breakage point verification means can designate a verification mode, and a damaged image region and A compressed image data restoration device characterized by changing a detection density at a boundary with an undamaged image region.   The compressed image data restoration device according to any one of claims 1 to 4, wherein the breakage point resetting means includes the breakage by the first breakage point verification means and the second breakage point verification means. A compressed image data restoration device which, when the compressed image data at the estimated point is changed, resets the damaged point if the restoration is impossible based on the changed compressed image data.   The compressed image data restoration device according to any one of claims 1 to 5, wherein the breakage point resetting unit sets the breakage estimation point so as to be close to the breakage estimation point and the original breakage point. A compressed image data restoration device characterized by being reset.   7. The compressed image data restoration device according to claim 1, wherein the compressed image data is a still image compressed image.   The compressed image data restoration device according to any one of claims 1 to 6, wherein the compressed image data is a moving image compressed image.   9. The compressed image data restoration apparatus according to claim 1, wherein the instruction operation unit replaces predetermined data if display is impossible when displaying damaged compressed image data. A compressed image data restoration device characterized in that it is possible to display.   The compressed image data restoration device according to any one of claims 1 to 9, wherein the instruction operation unit can detect and detect the presence or absence of an image breakage point when the compressed image data is restored and displayed. The compressed image data restoration device, wherein the instruction operation is accepted when the image data is received. An instruction operation process for restoring and displaying the compressed image data and receiving an instruction operation for an image breakage point;
An initial setting step for setting a damage estimation point in the compressed image data corresponding to the relative position information in the restored image data of the instructed image damage point;
The damage estimation point is detected by detecting a difference between the case where the compressed image data of the damage estimation point is changed and the restoration process is performed based on the compressed image data after the change and the restoration process is performed based on the original compression image data. A first failure point verification process for verifying,
Displacement of the image at the boundary between the damaged image area and the undamaged image area when the compressed image data at the estimated damage point is changed and restored based on the compressed image data after the change A second failure point verification step for verifying the damage estimation point based on the situation;
A failure point resetting step for controlling execution of the first breakage point verification step and the second breakage point verification step, and bringing the breakage estimation point closer to the original breakage point based on the verification result; A compressed image data restoration method characterized by the above. A recording medium storing a compressed image data restoration program for restoring compressed image data by a computer,
An instruction operation function that restores and displays the compressed image data and accepts an instruction operation for an image breakage point,
An initial setting function for setting a damage estimation point in the compressed image data corresponding to the relative position information in the restored image data of the instructed image damage point;
The damage estimation point is detected by detecting a difference between the case where the compressed image data of the damage estimation point is changed and the restoration process is performed based on the compressed image data after the change and the restoration process is performed based on the original compression image data. The first breakage point verification function to verify,
Displacement of the image at the boundary between the damaged image area and the undamaged image area when the compressed image data at the estimated damage point is changed and restored based on the compressed image data after the change A second breakage point verification function for verifying the breakage estimation point based on the situation;
A failure point reset function for controlling execution of the first breakage point verification function and the second breakage point verification function and bringing the breakage estimation point closer to the original breakage point based on the verification result; A medium on which a compressed image data restoration program is recorded.

Images