KR100528324B1 - Apparatus and method for coding and decoding image using overlapped rectangular slices - Google Patents

Abstract

Disclosed are an image encoding and decoding apparatus and method using overlapping rectangular slice structures. The apparatus for encoding an image may include a slice modeling unit classifying each image into at least one independent rectangular slice; A picture header encoding unit encoding position and size information of the slices classified by the slice modeling unit together with other information in a picture header, but omitting and encoding information repeated for each image; And a slice encoder for encoding an image in units of slices with reference to the picture header information. Here, at least two or more of the slices are configured such that predetermined regions overlap.

Description

Apparatus and method for coding and decoding image using overlapped rectangular slices}

The present invention relates to encoding and decoding of an image, and more particularly, to an apparatus and method for encoding and decoding an image using a rectangular slice structure.

The slice structure is useful when encoding an image and transmitting it through a network. In general, slices are decodable independently, so that if an error occurs during transmission, the error is limited to that slice. Therefore, the slice structure has a characteristic that is robust to errors.

Existing video coding standards include techniques for efficiently using this slice structure. For example, the ITU-T H.263 Recommendation provides for the use of rectangular slices in the Annex K (Slice Structured Mode).

Recently, ISO / IEC MPEG and ITU-T VCEG jointly formed a joint video team (JVT) to develop a new video coding standard, which uses a more advanced slice structure than the slice structure of H.263. This was adopted. This technique is called flexible MB ordering and allows you to organize slices in any macroblock (MB) order. In this case, the MB allocation map is used to define which macroblock is included in which slice. In the case of a rectangular slice, the index of the upper left macroblock and the lower right macroblock included in this slice are used. The index is used to define the position and size of the rectangular slice. The rectangular slices configured as described above not only have better compression efficiency than slices formed by columns of macro blocks, but also can implement various encoding techniques according to the position selection of the rectangular slices.

An object of the present invention is to configure a rectangular slice overlapping each other when the image is divided into slices and encoded, so that the boundary portion does not appear when encoding the ROI, and can support the image in the image of various sizes and To provide a method.

Another object of the present invention is to provide an apparatus and method capable of having a characteristic that is robust to errors in network transmission by encoding overlapping portions of slices.

Another object of the present invention is to provide an apparatus and a method for improving encoding efficiency by omitting information repeated for each image during encoding of a picture header.

Another object of the present invention is to provide a recording medium recorded with a program code executable by a computer in the encoding and decoding method using the above method.

In order to achieve the above object, an image encoding apparatus according to the present invention includes a slice modeling unit for classifying each image into at least one independent rectangular slice; A picture header encoding unit encoding position and size information of the slices classified by the slice modeling unit together with other information in a picture header, and omitting and encoding information repeated for each image; And a slice encoder configured to encode an image in units of slices with reference to the picture header information, wherein at least two or more of the slices are configured to overlap a predetermined region.

In order to achieve the above object, an image decoding apparatus according to the present invention includes: a picture header decoder for decoding a picture header in a bit string; A slice configuration unit constituting a slice through the position and size information of the slice in the picture header information; A slice decoder which decodes an image in units of slices with reference to the picture header; And an image configuration unit configured to configure the images of the slice unit reconstructed by the slice decoding unit into a single image by referring to the position and size information of the slice obtained by the slice configuration unit, and at least two of the reconstructed slices. The above is characterized in that the predetermined areas are configured to overlap.

In order to achieve the above object, the image encoding method according to the present invention comprises the steps of: (a) setting the position and size of a slice in a single image; (b) encoding the position and size information of the slice set in the step (a) together with other information in a picture header; And (c) encoding the image in units of slices with reference to the picture header information encoded in the step (b), wherein at least two or more of the slices are configured to overlap a predetermined region. do.

In order to achieve the above object, an image decoding method according to the present invention includes: (a) decoding a picture header in a bit string; (b) constructing a slice using location and size information of the slice included in the picture header decoded in step (a); (c) decoding an image in units of slices with reference to the picture header decoded in step (a); And (d) configuring images of the slice unit decoded in the step (c) into a single image by referring to the position and size information of the slice, wherein at least two of the slices are a predetermined region. It is characterized by being configured to overlap.

Hereinafter, with reference to the accompanying drawings will be described in detail a preferred embodiment of the present invention.

1 is a block diagram of an image encoding apparatus 100 according to a preferred embodiment of the present invention. Referring to FIG. 1, the image encoding apparatus 100 according to the present invention includes a slice modeling unit 120, a picture header encoding unit 130, and a slice encoding unit 140. The slice encoder 140 includes a space / temporal prediction encoder 141, a transform and quantizer 142, and an entropy encoder 143.

The slice modeling unit 120 classifies each image 110 into at least one independent slice so that a user can independently encode any desired region. In this case, the classified slice regions may have a quadrangular shape, and each rectangular region may overlap each other. The slice regions represented by the rectangles do not necessarily represent the entire image, and portions not displayed in one slice region are divided into another independent slice. The encoding method for the case where the rectangular slices overlap will be described in detail below with reference to FIGS. 5 and 6.

The picture header encoder 130 encodes common information required to decode all slices existing in one image, and transmits the encoded information to the slice encoder 140. Information transmitted at this time includes the number, shape, position and size of the slice. The information in the picture header may be encoded with a variable length code or may be encoded with a fixed length code. The picture header encoder may insert redundant encoding indication information into the picture header in order to determine whether to encode the overlapped portions at the time of encoding rectangular slices partially or completely overlapping a predetermined region.

In the picture header configured as described above, the encoding of the picture header is omitted since information for overlapping information used in every video is not encoded each time and the overlapping portion is omitted. This is done by indicating that some information in the picture header of the current picture is the same as the information in the picture header of the previous picture. This will be described in detail below with reference to FIG. 7.

The slice encoder 140 encodes an image in units of slices by referring to picture header information input from the picture header encoder 130. To this end, the space / time prediction encoder 141 removes overlapping information in space / time. The transform and quantization unit 142 performs a discrete cosine transform (DCT) transform on the output of the space / temporal prediction encoder 141 and quantizes the transform coefficients. The entropy encoder 143 entropy encodes the output of the transform and quantizer 142 to generate a compressed bit string.

The slice encoder 140 encodes an image by encoding the image by dividing it into slice units when the image is transmitted through a network. By dividing and encoding the image by the rectangular slice unit, prediction slice loss between slices is reduced, ROI encoding and intra-image encoding. And so on. Here, the ROI encoding is an encoding method that improves subjective image quality under a limited bit rate by dividing an image into an ROI and a background region to improve the image quality of the ROI and the image quality of the background region. In addition, the intra-image encoding is a method of independently decoding a portion composed of rectangular slices so that the portion can be used like another image.

In the present invention, when the regions represented by the rectangular slices overlap each other, a predetermined region existing between the overlapping portion and the background region is set as the transition region, thereby preventing a problem of deterioration in image quality between the ROI and the background region during encoding. In addition, by overlapping the overlapping portions of the slice regions, not only are they robust to errors in network transmission, but also images of various sizes can be used when encoding images in the image.

2 is a flowchart illustrating an image encoding method of the image encoding apparatus 100 illustrated in FIG. 1. Referring to FIG. 2, when the position and size of a slice are first set, the slice modeling unit 120 classifies the corresponding image into at least one or more independent slices (step 1200). The picture header is encoded by the picture header encoder 130 (step 1300), and the slices are encoded by the slice encoder 140 (step 1400).

Here, the slice encoding step (step 1400) may include a space / temporal prediction encoding step (step 1410) of removing space / time overlapping information existing in a slice unit image, and discrete cosine for the data from which duplicate information is removed. Performing transform (DCT) transform and quantization (step 1420). And entropy encoding (step 1430) of entropy encoding the quantized data to generate a compressed bit string.

3 is a block diagram of an image decoding apparatus 200 according to a preferred embodiment of the present invention. Referring to FIG. 3, the image decoding apparatus 200 according to the present invention includes a picture header decoder 250, a slice composer 260, a slice decoder 270, and an image composer 280. The slice decoder 270 includes an entropy decoder 271, an inverse quantization and inverse transform unit 272, and an image reconstruction unit 273.

The picture header decoder 250 decodes picture header information from a bit string received through a network. In addition, information about the number, shape, position, and size of the decoded slice is transmitted to the slice configuration unit 260, and other information is transmitted to the slice decoding unit 270.

The slice configuration unit 260 configures slices by selecting slice positions and processing overlapping portions of slices in response to information about the number, shape, position, and size of the slices received from the picture header decoder 250. do. Processing for the portion where the slice regions overlap will be described in detail below with reference to FIGS. 5 and 6.

The slice decoder 270 decodes an image of each slice unit by referring to picture header information input from the picture header decoder 250. To this end, the entropy decoding unit 271 entropy decodes the bit string, and the inverse quantization and inverse transform unit 272 performs inverse quantization and inverse DCT transformation on the entropy decoded bit string. The image reconstructor 273 reconstructs the image by compensating spatial / temporal predictive encoded information on the output data of the inverse quantization and inverse transform unit 272. At this time, the image reconstructed in each slice unit is merged into one image by the image constituting unit 280 according to the information input from the slice constituting unit 260.

The video decoding apparatus 200 according to the present invention having such a configuration is for decoding an image encoded by the video encoding apparatus 100 shown in FIG. 1, and the operation procedure thereof is the reverse of that of the video encoding apparatus 100. However, the basic characteristics for the slice processing that the image decoding apparatus 200 has are the same as those of the image encoding apparatus 100. Therefore, in order to simplify the description, overlapping descriptions of characteristics common to the image encoding apparatus 100 will be omitted.

4 is a flowchart illustrating an image decoding method of the image decoding apparatus 200 illustrated in FIG. 3. Referring to FIG. 4, picture header information is first decoded in the received bit string (step 2500). At this time, information about the number, shape, position, and size of the slices is transmitted to the slice constituting unit 260, and the slice constituting unit 260 selects the positions of the slices in response to the input information, and overlaps the slices. The portion is processed to form a slice (step 2600). Then, decoding of the configured slice is performed (step 2700), and an image is constructed by referring to the position and size of the slice (step 2800).

Here, the slice decoding step 2700 may include an entropy decoding step 2710 for entropy decoding a bit string, an inverse quantization and an inverse DCT transform on the entropy decoded data, and an inverse DCT transform. Compensating the spatial / temporal predictive coded information on the extracted data (step 2730).

FIG. 5 is a diagram for describing a method of processing the slice when one rectangular slice including the ROI in one image is completely included in another rectangular slice. Referring to FIG. 5, one image 300 includes two rectangular slices 302 and 303 and one background slice 304.

For the two rectangular slices 302 and 303, the small rectangular slice 302 is completely contained within the area of the large rectangular slice 303. In this case, the area of the large rectangular slice 303 may overlap the area indicated by the small rectangular slice 302 (part indicated by hatching), and the small rectangular slice so that the two rectangular slices 302 and 303 do not overlap each other. An area excluding 302 (an area excluding the hatched portion of the area of the large rectangular slice 303) may be shown. As such, when the regions represented by the two slices 302 and 303 do not overlap each other, the small rectangular slice 302 is a portion of the image that the user considers more important than the view, that is, the region of interest 301. The large rectangular slice 303 is used as a slice including a transition region located between the region of interest 301 and the background region 304.

In general, an encoding apparatus using a rectangular slice improves the image quality of the ROI when encoding the ROI and reduces the image quality by compressing the background region much, thereby increasing the subjective image quality under a limited bit amount. However, at this time, a boundary is formed between the ROI 301 and the background region 304 due to a sudden change in image quality between the ROIs. The encoding apparatus 100 according to the present invention may have the ROI 301 as shown in FIG. 5. By setting the transition region 303 between the < RTI ID = 0.0 > and the background region 304, this boundary generation can be effectively reduced.

Here, since the small rectangular slice 302 can be independently decoded, when it is impossible or unnecessary to decode a single whole image, only the small rectangular slice region including the ROI is decoded and used. At this time, the image reconstructed from the small rectangular slice 302 becomes an image in the image. When the area represented by the large rectangular slice 303 does not overlap with the area represented by the small rectangular slice 302 (that is, the large rectangular slice 303 is a transition region between the ROI 301 and the background region 304). If used as 303), the image reconstructed from the rectangular slice 303 forms another image in the image. Therefore, the number of square slices in a single image can be obtained step by step, that is, from small images to larger and larger images.

On the other hand, when the portions indicated by the two rectangular slices 302 and 303 overlap (that is, when the large rectangular slice 303 completely includes the small rectangular slice 302), the portion indicated by the small rectangular slice 302 (Ie, hatched portions) are encoded and decoded again by the large rectangular slice 303. This method is useful when used in a faulty network. Because the area indicated by the small rectangular slice 302 is considered to be important as the area of interest 301, another part of the image which is not an error when an error occurs in any of the slices transmitted after encoding and transmitting this part redundantly This is because by using, it is possible to prevent an error that may occur during network transmission. In this case, the two rectangular slices 302 and 303 constitute the independent intra-images as described above.

FIG. 6 is a diagram for describing a method of processing the slice when two different rectangular slices including the ROI overlap each other without being included in one image.

Referring to FIG. 6, two rectangular slices 312 and 313 both include a region of interest 311. Even in this case, as in the case of FIG. 5, the overlapping area (that is, the hatched area) represented by the two rectangular slices is overlapped with the case where the non-overlapping area (ie, the area of any slice is hatched). If it consists of areas other than)). When the areas represented by the two rectangular slices overlap each other, as in the case of FIG. 5, since the common areas are overlapped and encoded, the network has strong characteristics against errors in network transmission. At this time, each slice forms an image in an independent image as described above.

Whether or not each slice is configured to include overlapping regions (hatched regions) between two slices can be indicated by inserting additional information into a slice header or a picture header. In addition, if one slice is included in another slice as shown in FIG. 5 without additional information, a rule may be determined such that the slice is not encoded when overlapped without overlapping as shown in FIG. 6. If there is priority information between slices, if the priorities are the same, they are encoded repeatedly, otherwise they may not overlap. Following the slice encoding as described above, the header encoding method according to the present invention will be described.

As described above, information such as the number, shape, position, and size of slices in one image is encoded in a picture header of the image. In addition, the picture header includes common information necessary for encoding a single video. Such information may change in every image when the video is encoded, but may have little information. Therefore, in the present invention, the information that does not change is not encoded for every picture but is taken from the picture header of the previous picture as it is, thereby improving coding efficiency.

In particular, if all MBs are encoded for each MB, such as the flexible MB ordering technique used in the JVT standard, the encoding amount of the picture header is increased. Accordingly, in the present invention, the same macroblock allocation map is used for each image, but the encoding efficiency is improved by using the picture header of the previous image as it is without transmitting the information each time.

7 is a diagram for describing a method of encoding a picture header according to an embodiment of the present invention. 7 illustrates a picture header corresponding to each image when two images are encoded. Here, picture header 1 means a picture header of a previous picture, and picture header 2 means a picture header of a current picture.

Referring to FIG. 7, each picture header has a portion that changes each time for each image, which is indicated by an A portion 400 of picture header 1 and an A ′ portion 410 of picture header 2. There is a part which hardly changes between the picture headers, which is indicated by the B part 401 of the picture header 1 and the B part 411 of the picture header 2. FIG. The portion of the picture header information that may or may not change is denoted by C 402 and C ′ 412. Here, the B portions 401 and 411 of each picture header represent substantially the same information.

In this case, the picture header encoding unit 130 of the image encoding apparatus 100 according to the present invention omits and encodes the same information as the B portions 401 and 411 of each picture header. That is, the B part 401 of the picture header 1, which is a part common between the picture headers, is omitted when the picture header 2 is encoded, and the D part 431 is added to indicate that the B part 401 is omitted.

For example, when the D portion 431 is composed of 1 bit, if the B portions 401 and 411 of the two picture headers are the same information, the D portion 431 displays a value of "1" to indicate the B portion 401. Is omitted. If the B portions 401 and 411 of the two picture headers are not the same information, " 0 " is displayed to encode the B portion 401 without omission.

In addition, when the D portion 431 includes two bits, not only the B portion 401 of the picture header 1 but also the C portion 402 may be omitted. For example, when the B portions 401 and 411 and the C portions 402 and 412 of the two picture headers each have the same information, the B portion 401 and the C portion are indicated by marking the D portion 431 as "11". Omit all of (402), and if only one is duplicated, mark it as "01" or "10" to omit only the duplicated part, and if both parts are not identical, mark as "00" and both parts Encode

As such, the picture header encoder 130 according to the present invention can increase coding efficiency by omitting and encoding the same information in the picture header. In order to omit the same information existing on the header, the picture header encoder 130 divides the picture header into certain parts and encodes the picture header by including information indicating whether any part is omitted.

In the above, the method of encoding and decoding by overlapping rectangular slice structures when encoding an image into a slice structure and the method of encoding a header of the image have been described above. The present invention can be usefully used when the image is encoded in a slice structure and transmitted through a network. This means that subjective image quality can be improved by dividing the region of interest and the background region differentially by using the visual characteristics of a person when dividing the image into slices, and in particular, by setting a transition region between the region of interest and the background region, the boundary appears. The phenomenon can be reduced. In addition, by overlapping the overlapping parts, important parts can be made more robust to errors.

The invention can also be embodied as computer readable code on a computer readable recording medium. The computer-readable recording medium includes all kinds of recording devices in which data that can be read by a computer system is stored. Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage, and the like, and may also be implemented in the form of a carrier wave (for example, transmission over the Internet). Include. The computer readable recording medium can also be distributed over network coupled computer systems so that the computer readable code is stored and executed in a distributed fashion.

As described above, according to the video encoding and decoding apparatus and method according to the present invention, the rectangular slices are overlapped with each other when the video is divided into slices and encoded and decoded, so that the subjective picture quality can be increased, and the region of interest and By setting the transition region between the background regions, the phenomenon in which the boundary appears can be reduced. In addition, by overlapping the overlapping parts, it is possible to be robust against errors in network transmission. When the picture header is encoded, the encoding information of the picture header can be improved by omitting and encoding information repeated for each image.

1 is a block diagram of a video encoding apparatus according to a preferred embodiment of the present invention.

FIG. 2 is a flowchart illustrating a video encoding method of the video encoding apparatus shown in FIG. 1.

3 is a block diagram of an image decoding apparatus according to a preferred embodiment of the present invention.

4 is a flowchart illustrating an image decoding method of the image decoding apparatus of FIG. 3.

FIG. 5 is a diagram for describing a method of processing the slice when one rectangular slice including the ROI in one image is completely included in another rectangular slice.

FIG. 6 is a diagram for describing a method of processing the slice when two different rectangular slices including the ROI overlap each other without being included in one image.

7 is a diagram for describing a method of encoding a picture header according to an embodiment of the present invention.

<Description of Symbols for Main Parts of Drawings>

100: video encoding apparatus 110, 290: video

120: slice modeling unit 130: picture header encoder

140: slice encoder 200: video decoding apparatus

250: picture header decoding unit 260: slice construction unit

270: slice decoder 280: video component

Claims (52)

A slice modeling unit classifying each image into at least one independent rectangular slice; A picture header encoding unit encoding position and size information of the slices classified by the slice modeling unit together with other information in a picture header, and omitting and encoding information repeated for each image; And A slice encoder which encodes an image in units of slices with reference to the picture header information; At least two or more of the slices are configured such that a predetermined area overlaps. The slice encoder of claim 1, wherein the slice encoder A spatial / temporal predictive encoding unit which performs spatial / temporal predictive encoding on the image in units of slices; A transform and quantizer for performing discrete cosine transform and quantization on the information encoded by the space / temporal prediction encoder; And And an entropy encoding unit for entropy encoding information obtained by the transform and quantization unit. The method of claim 1, The slice encoder encodes the image quality of the ROI that the user considers important in image quality and encodes the background region of the image excluding the ROI to have low image quality, thereby improving subjective image quality under a limited bit rate. An image encoding device. The method of claim 3, wherein When the two slices overlap each other such that one rectangular slice is completely included in another rectangular slice, the slice encoder may include an inner slice region included inside of the slices, and an outer slice region between the outer slice region and the inner slice region. An image encoding device, characterized in that for each encoding independently. The method of claim 4, wherein And a region between the outer slice region and the inner slice region is a transition region that buffers a boundary between the inner slice region and the background region during encoding. The method of claim 4, wherein When the two slices overlap each other so that one rectangular slice is completely included in another rectangular slice, the slice encoding unit encodes the overlapped regions for each slice so as to be robust to errors in network transmission. Video encoding device. The method of claim 1, The slice encoder, when one rectangular slice overlaps with a part of another rectangular slice, encodes the overlapped region for each slice. The method according to claim 4 or 7, And the slice encoder is configured to encode each rectangular slice so as to be used as an image in the image. The method according to claim 4 or 7, The picture header encoding unit inserts redundant encoding indication information into the picture header to determine whether to encode the overlapped portions at the time of encoding rectangular slices partially or completely overlapping a predetermined region. Video encoding device. The method of claim 9, The slice encoder determines whether to overlap the overlapped portions by using the overlapped encoding indication information or the priority between the slices when the overlapping rectangular slices are encoded. An image encoding device, characterized in that the encoding is repeated. The method of claim 9, And the slice encoder determines whether to overlap the overlapped portions according to whether or not the rectangular slices are completely overlapped instead of the overlapped encoding indication information. The method of claim 1, And the picture header encoding unit encodes the picture header of the current video by omitting information overlapping with the picture header of the previous video. The method of claim 12, The picture header encoding unit divides the picture header into at least two or more parts according to the degree of frequent change in each image, and encodes information in the picture header indicating whether the part is omitted. . A picture header decoder for decoding a picture header from a bit string; A slice configuration unit constituting a slice through the position and size information of the slice in the picture header information; A slice decoder which decodes an image in units of slices with reference to the picture header; And An image construction unit configured to configure the images of a slice unit reconstructed by the slice decoding unit into a single image by referring to the position and size information of the slice obtained from the slice construction unit; At least two or more of the reconstructed slices are configured such that a predetermined area overlaps. 15. The method of claim 14, wherein the slice decoder An entropy decoding unit for entropy decoding the bit string; An inverse quantization and inverse transform unit for inverse quantization and inverse discrete cosine transforming information decoded by the entropy decoding unit; And And an image reconstruction unit for reconstructing the image in units of slices by performing temporal / spatial prediction compensation on the information obtained by the inverse quantization and inverse transform unit. The method of claim 14, The slice decoder may decode the ROI that the user regards as important in image quality, and decode the background region excluding the ROI in the image so that the subjective image quality is improved under a limited bit rate. An image decoding device. The method of claim 16, When the two slices overlap each other such that one rectangular slice is completely included in another rectangular slice, the slice decoder includes an internal slice region included inside of the slices, and between the external slice region and the internal slice region. And a respective decoding region independently. The method of claim 17, And a region between the outer slice region and the inner slice region is a transition region that buffers the boundary between the inner slice region and the background region when decoding. The method of claim 17, When the two slices overlap so that one rectangular slice is completely included in the other rectangular slice, the slice decoder is robust to errors by decoding data in which no error occurs among data encoded in the overlapped region. And a video decoding apparatus. The method of claim 14, The slice decoder, when one rectangular slice overlaps with a portion of another rectangular slice, decodes the overlapped region for each slice. The method of claim 17 or 20, And the slice decoder decodes each square slice so that each square slice can be used as an image in the image. The method of claim 17 or 20, The picture header decoder decodes duplicated decoding indication information in the picture header to determine whether to decode the overlapped portion repeatedly when decoding rectangular slices partially or completely overlapping a predetermined region. An image decoding device. The method of claim 22, The slice decoder determines whether to overlap the decoded portion by using the duplicated decoding indication information or the priority between the slices when decoding the overlapping rectangular slices. An image decoding device characterized in that the decoding repeatedly. The method of claim 22, And the slice decoder determines whether to overlap and decode the overlapped portions according to whether or not the rectangular slices are completely overlapped instead of the duplicated decoding indication information. The method of claim 14, And the picture header decoding unit decodes the picture header of the current video using information overlapping with the picture header of the previous video as it is. The method of claim 25, The picture header decoder may decode information indicating that a predetermined portion is omitted from a picture header of the current video, and use the omitted information from the picture header of the previous video in response to the decoded information. An image decoding device. (a) setting the position and size of the slice in one image; (b) encoding the position and size information of the slice set in the step (a) together with other information in a picture header; And (c) encoding the image in units of slices by referring to the picture header information encoded in the step (b), At least two or more of the slices are configured such that a predetermined area overlaps. The method of claim 27, In the step (b), encoding is performed by omitting information repeated for each image. The method of claim 28, In the step (b), the picture header is divided into at least two or more parts according to the degree of frequent change in each image, and image encoding is performed by encoding information indicating whether the part is omitted. Way. The method of claim 27, In the step (b), redundant encoding indication information is inserted into the picture header in order to determine whether to encode the overlapped portions at the time of encoding rectangular slices partially or completely overlapping a predetermined region. The video encoding method. The method of claim 27, wherein step (c) (c-1) performing spatial / temporal prediction coding on the image in units of slices; (c-2) performing discrete cosine transform and quantization on the information predicted and encoded in step (c-1); And (c-3) entropy encoding the information obtained in the step (c-2). The method of claim 27, In the step (c), the region of interest, which is considered to be important to the user, is encoded with good image quality, and the region of the image except the region of interest is encoded with poor image quality, thereby improving subjective image quality under a limited bit rate. An image encoding method. The method of claim 32, In the step (c), when the two slices overlap such that one rectangular slice is completely included in the other rectangular slice, an inner slice region included inside of the slices, an outer slice region, and an inner slice included outside An image encoding method, characterized in that each region between regions is encoded independently. The method of claim 33, wherein And a region between the outer slice region and the inner slice region is a transition region that buffers a boundary between the inner slice region and the background region during encoding. The method of claim 33, wherein In the step (c), when the two slices overlap each other such that one rectangular slice is completely included in another rectangular slice, the overlapped regions are encoded by overlapping each slice, thereby making it robust to errors in network transmission. A video encoding method. The method of claim 27, In the step (c), when one rectangular slice overlaps with a part of another rectangular slice, the overlapping region is encoded by overlapping each slice. The method of claim 33 or 36, In the step (c), the respective rectangular slices are encoded such that they can be used as images in the image. The method of claim 27 or 30, In the step (c), when the overlapping rectangular slices are encoded, the overlapped encoding mark information or the priority between the slices is used to determine whether to overlap the overlapped portions. If the priorities are the same, the overlapped portions are overlapped. The video encoding method, characterized by overlapping portions. The method of claim 27 or 30, In the step (c), it is determined whether or not the overlapped portions are to be encoded by overlapping the rectangular slices according to whether the rectangular slices are completely overlapped instead of the overlapped encoding indication information. (a) decoding a picture header in a bit string; (b) constructing a slice using location and size information of the slice included in the picture header decoded in step (a); (c) decoding an image in units of slices with reference to the picture header decoded in step (a); And (d) composing the images of the slice unit decoded in the step (c) into a single image by referring to the position and size information of the slice, At least two of the slices are configured such that a predetermined area overlaps. The method of claim 40, Step (b) is to decode the information indicating that a portion of the picture header of the current picture is omitted, and to use the omitted information from the picture header of the previous picture in response to the decoded information Image Decoding Method. The method of claim 40, In the step (b), the redundant decoding indication information is decoded in the picture header in order to determine whether to decode the overlapped portion at the time of decoding the rectangular slices partially or completely overlapping a predetermined region. An image decoding method. 41. The method of claim 40, wherein step (c) (c-1) entropy decoding the bit string; (c-2) performing inverse quantization and inverse discrete cosine transform on the information decoded in step (c-1); And and (c-3) restoring the image in units of slices by performing temporal / spatial prediction compensation on the information obtained in step (c-2). The method of claim 40, In the step (c), the region of interest, which is considered to be important to the user, is decoded with good image quality, and the background region other than the region of interest is decoded with poor image quality, thereby improving subjective image quality under a limited bit rate. An image decoding method characterized by. The method of claim 44, In the step (c), when the two slices overlap such that one rectangular slice is completely included in the other rectangular slice, an inner slice region included inside of the slices, an outer slice region, and an inner slice included outside An image decoding method characterized by independently decoding regions between regions. The method of claim 45, And a region between the outer slice region and the inner slice region is a transition region that buffers a boundary between the inner slice region and the background region during decoding. The method of claim 45, In the step (c), when the two slices overlap so that one rectangular slice is completely included in the other rectangular slice, the error is decoded by decoding the data in which no error is generated among the overlapped encoded data for the overlapped region. The image decoding method characterized in that the robust to. The method of claim 40, In the step (c), when one rectangular slice overlaps with a portion of another rectangular slice, the overlapped region is repeatedly decoded for each slice. 49. The method of claim 45 or 48, In the step (c), each of the rectangular slices is decoded so that each can be used as an image in an image. 43. The method of claim 40 or 42 wherein In the step (c), when the overlapping rectangular slices are decoded, the duplicated decoding indication information or the priority between the slices is used to determine whether to decode the overlapped portions. If the priorities are the same, the overlapping portions are overlapped. An image decoding method, characterized in that the portion is decoded in duplicate. 43. The method of claim 40 or 42 wherein In the step (c), it is determined whether or not the overlapped portions are to be repeatedly decoded according to whether the rectangular slices are completely overlapped instead of the duplicated decoding indication information. A computer-readable recording medium having recorded thereon a program for executing the method of claim 27 on a computer.