JPH01181280A - Coding system for picture signal - Google Patents

Abstract

PURPOSE:To attain local display of a coded picture and processing without complicated procedure or long processing by dividing a picture into plural partial pictures, coding and storing them. CONSTITUTION:A picture signal storage section 1 converts a picture signal by one pattern outputted from a picture input device such as a television camera into a digital quantity and stores it as a picture signal given from a matrix picture element. An optional shape area setting section 2 reads a picture signal form the picture signal storage section 1 and sets an optional shaped area in the picture. When the setting of all optional shaped areas in the picture is finished, the stored said shape matrix is sent sequentially to a rectangular area setting section 3. A rectangular area coding data generated by a rectangular area coding section 5 is sent to a code storage section 6 together with rectangular area location information. Thus, the partial display or processing of the coded picture element is applied with less processing time without complicated processing.

Description

DETAILED DESCRIPTION OF THE INVENTION (Technical field to which the invention pertains) The present invention relates to an image signal encoding method for digitizing and encoding an image signal output from an image input device such as a camera.

(Prior art) Conventionally, in the encoding method of image signals output from an image input device such as a camera, - images are decomposed into a matrix of minute pixels, and the image signal for each pixel is converted into a digital quantity. , and encode all pixels at once as a set of codes for each pixel.

Since - images are encoded all at once, - images cannot be treated as a set of data, so processing of one image is required. When extracting only a partial image and replaying it or compositing that partial image with another image, a person must specify the area to be processed in detail each time, or automatic processing such as edge extraction is required. It is necessary to apply an identification method, which causes problems in that the procedure becomes complicated and the processing time becomes long.

(Objective of the Invention) The present invention does not encode - images all at once, but divides them into multiple partial images and stores them.
The object of the present invention is to realize an image signal encoding method that enables partial display and processing of encoded images without complicated procedures or extensive processing.

(Structure of the Invention) (Characteristics of the Invention and Differences from the Prior Art) The present invention incorporates an image signal of one image into an arbitrary-shaped region that is a closed region and a rectangular region that includes the region. This is set as a partial image, and multiple partial images are provided within the image to cover all areas within the image. Each partial image is encoded independently, and furthermore, the The most important feature is that one image is encoded by simultaneously encoding position information.

This is different from the conventional method of encoding and storing one image signal as one piece of data.

(Embodiment) FIG. 1 shows a block diagram of the configuration of the first embodiment of the present invention, in which 1 is an image signal storage unit that temporarily stores an input image signal, and 2 is an image signal storage unit that temporarily stores an input image signal. 3 is a rectangular area setting unit that sets a rectangular area that includes the arbitrarily shaped area; 4 is an input image signal; 5 is a dummy signal generator that generates an irrelevant dummy signal; 5 is a rectangular area encoder that encodes the inside of a rectangular area; 6 is a rectangular area encoder that encodes the inside of a rectangular area;
is a code volume unit that stores encoded output as image data.

An image signal is a luminance signal in the case of a monochrome image, a luminance signal of R (red) component, G (green) component, and B (blue) component in the case of a color image, or Y (luminance signal) in the NT and SC systems. component, IQ (chromaticity signal) component. In the following explanation, for convenience, only the case of a monochrome image will be described, but the case of a color image can also be realized by applying the operations for a monochrome image in parallel or repeatedly to each component of the color image. For.

Due to the structure of the present invention, either a black and white image or a color image may be used.

Next, the operation of the first embodiment will be explained using FIGS. 2 to 6.

The image signal storage unit 1 converts an image signal for one screen outputted from an image input device such as a television camera into a digital quantity, and as shown in the example of an input image in FIG.
It is stored as an image signal given to each pixel in a matrix of pixels (H and V are positive integers). In FIG. 2, closed curves 101, 102, and 103 are shown for use in later explanations, and are not signals stored in the image signal storage section 1.

The arbitrary shape area setting section 2 reads out the image signal from the image signal storage section 1 and sets an arbitrary shape area within the image. Here, the arbitrarily shaped area is a closed area that is a part of an image, that is, an area surrounded by a closed curve. For example, in FIG. 2, there is a region 111 inside the closed curve 101, a region 112 inside the closed curve 102, a region 112 inside the closed curve 103, and a closed curve lot and a closed curve 1.
Area 113 outside of 02 is an example of an arbitrarily shaped area in the image of FIG.

The arbitrary shape area setting section 2 has the following three area setting modes, and operates by switching the area setting mode using a switch. What are the three area setting modes?

Area setting mode: Talk with the operator and set the area as defined by the closed curve manually drawn by the operator as a figure.

Area setting mode 2: Regardless of the content of the image, before starting the operation, the area is set in advance in the arbitrarily shaped area setting section 2 according to the area stored in the form of a shape matrix to be described later.

Region setting mode 3: A known image processing method such as an edge extraction method is applied, and the region is set as defined by the extracted closed curve.

It is.

The arbitrary shape area is set by converting the read image signal into a shape matrix. What is a shape matrix?
There is a one-to-one correspondence with each pixel. In a matrix consisting of horizontal H2 vertical V elements, if the corresponding pixel is included in the setting area, a value of 1 is assigned to the element; otherwise, a value of 0 is assigned to the element. It is. However, when area setting mode 2 is selected, the already set shape matrix is copied as is.

FIGS. 3(a), (b), and (c) respectively show examples of shape matrices related to arbitrary shape regions 111, 112, and 113 in FIG. 2. However, the closed curves 101, 102, and 103 in FIG. 3 are displayed to make the diagram easier to understand, and are not elements of the shape 71-lix.

The window shape region setting unit 2 can set a plurality of arbitrary shape regions (111.-112.113 in FIG. It is assumed that the data can be stored in figures (a), (b), and (c)). When all arbitrary shape areas in the image have been set, the stored shape matrices are sequentially sent to the rectangular area setting section 3. In the following description of the operation, only the operation related to one shape matrix will be described to simplify the explanation.

The rectangular area setting section 3 sets a rectangular area based on the sent shape matrix. The rectangular area is a rectangular area that includes an arbitrarily shaped area. Setting of a rectangular area is performed by creating a rectangular area matrix and rectangular area position information in the following steps (a) to (f).

(From the 0.1 pattern of the shape matrix shown in Fig. 3, M pixels horizontally and N pixels vertically (M≦H
, N≦V) is detected.

(0) From a shape matrix consisting of HXV elements, MXN corresponds one-to-one to each pixel in the detected rectangular area.
A partial matrix consisting of elements is extracted and used as a rectangular region matrix.

FIGS. 4(a), (b), and (c) respectively show examples of shape matrices related to arbitrary shape regions 111, 112, and 113 in FIG. 2.

(C) As illustrated in Figure 5, the origin is the top left corner of the original image, and the two sides that intersect with this are
When the axis is the y-axis, the coordinates (X,

ya) and the coordinates (Xb, Vh) of the apex of the lower right corner are detected and used as rectangular area position information. The value of the coordinates is
For example, it is assumed that it is represented by (row number 9 column number) of the shape matrix.

The above-mentioned rectangular area matrix (FIG. 4) and rectangular area position information (FIG. 5) are sent to the rectangular area encoding section 5. This rectangular area encoding unit 5 checks the values of the elements of the rectangular area matrix (FIG. 4) sent from the rectangular area setting unit 3, and if the value is O, transmits a dummy signal to the dummy signal generating unit 4. If the value is 1, the image signal of the corresponding pixel is requested from the image signal storage unit 1, and the received signal is encoded. This operation is performed for all elements of the rectangular area matrix to create a code string, which is used as rectangular area encoded data.

Through the above operations, encoding of only a part of the image signal according to claim (1) is realized.

The rectangular area encoded data created by the rectangular area encoder 5 in the previous operations is sent to the code storage unit 6 together with the rectangular area position information.

The code storage unit 6 assigns a serial number to each rectangular area on a first-come, first-served basis, and uses this as an index (FIG. 6 (A)). Vertex coordinates, (lower right vertex coordinates) storage format, and rectangular area encoded data is
It is stored in the storage format of rectangular area encoded data (1) illustrated in the figure. When the encoding storage unit 6 completes the storage operation, it sends a completion signal to the arbitrary shape area setting unit 2. By performing the above operations for all the shape matrices stored in the arbitrary shape region setting section 2, claim (3) is achieved.
encoding of image signals is realized.

FIG. 8 shows a block diagram of the configuration of a second embodiment of the present invention that encodes only a part of an input image.
In this embodiment, the dummy signal generating section 4 in the configuration of the embodiment is removed, and the rectangular area encoding section 5 is divided into an intra-area encoding section 51 and a shape encoding section 52. The other parts are the same as the configuration of the first embodiment.

Next, the operation of the second embodiment will be explained using FIG. 9.

However, in the second embodiment, the image signal storage unit 1. Arbitrary shape area setting section 2. Since the operation of the rectangular area setting section 3 is the same as that of the first embodiment, only the operations of the rectangular area encoding section 5 and the encoded storage section 6 will be explained to avoid duplication of explanation.

The intra-area encoding unit 51 checks the values of the elements of the rectangular area matrix among the rectangular area matrix (FIG. 4) and the rectangular area position information (FIG. 5) sent from the rectangular area setting unit 3, and calculates the value. When the value is 1, the image signal of the corresponding pixel is read from the image signal storage section 1 and encoded, and when the value is 0, no code is generated. This operation is performed for all elements of the rectangular area matrix to create a code string, which is used as encoded data in an arbitrary shape area (FIG. 9 (step □)).

The encoded data within the arbitrary shape region, the rectangular region matrix, and the rectangular region position information are sent to the shape encoding section 52.

The shape encoding unit 52 examines the values of the elements of the rectangular region matrix (FIG. 4), and generates arbitrary region shape data in which the element numbers of the elements whose value is 1 adjacent to the element whose value is 0 are listed. (Figure 9 (Step 2)) is created. The element number is
As mentioned above, the element at the upper left corner of the rectangular area matrix is represented by (0, O), the element at the lower right corner is represented by (M, N), and so on (row number 9 column number).

With the above configuration and operation, encoding of an image signal according to claim (2) is realized.

Arbitrarily shaped region encoded data (Fig. 9 (Step □)), arbitrary region shape data (Fig. 9 (Step 2)), and rectangular region position information generated by the rectangular region encoding unit 5 in the operations up to now (FIG. 5) is sent to the code storage section 6.

The code storage unit 6 assigns a serial number to each rectangular area on a first-come, first-served basis, and uses this as an index to store rectangular area position information in the area position information storage format illustrated in FIG. The intra-area encoded data is stored in the rectangular area encoded data storage format illustrated in FIG. When the code storage unit 6 completes the storage operation.

A completion signal is sent to the arbitrary shape region setting section 2. By performing the above operations for all shape matrices stored in the arbitrary shape region setting section 2.

Image signal encoding according to claim (3) is realized in which the entire input image is encoded.

FIG. 10 shows a configuration block diagram of the data compression section 7, which is inserted between the rectangular area encoding section 5 and the code storage section 6 in the first embodiment or the second embodiment. This constitutes the third embodiment. The data compression unit 7 includes a compression encoding unit 71 that performs compression encoding using a known image signal compression encoding method, and a compression encoding unit 72 that performs compression encoding using another known image signal compression encoding method. , a selection control section 70 that selects and operates one compression encoding section from the plurality of compression encoding sections.

In this embodiment, there are two compression encoding units, but the number is not limited to two, and three or more may be provided as necessary. Further, even in the same compression encoding method, different compression encoding units with different values of parameters controlling the degree of compression etc. may be set as separate compression encoding units. 7
Except for the insertion of , the other parts are the same as the first embodiment (FIG. 1) or the second embodiment (FIG. 8).

Next, the operation of the third embodiment will be explained using FIG. 11.

However, the operations of the one-image signal storage unit 1, arbitrary shape area setting unit 2, rectangular area setting unit 3, dummy signal generation unit 4, and rectangular area encoding unit 5 in the third embodiment are different from those in the first embodiment or the rectangular area encoding unit 5. Since the operation is the same as that of the second embodiment, only the operations of the data compression section 7 and the code storage section 6 will be explained to avoid duplication of explanation.

The data compression unit 7 receives data related to the rectangular area sent from the rectangular encoding unit 6 (in the first embodiment, rectangular area position information (FIG. 5) and rectangular area encoded data (FIG. 7), In the second embodiment, rectangular area position information (
Figure 6) and encoded data in arbitrary shape regions (Figure 9 (Eng.,
)) and arbitrary area shape data (Fig. 9 (Eng. 2))), compression encoding is [Recent trends in encoding technology (Journal of Television Society, Vol. 39. No.
10. This is a coding method typified by run-length coding, orthogonal transform coding, etc. described in ``pp.853-860.1985'', and realizes coding using the code amount of the original data.

In the configuration of the present invention, compression encoding is performed in a compression encoding section 71 or a compression encoding section 72.

The selection control unit 70 selects a compression encoding method each time data regarding one rectangular area is sent from the rectangular area encoding unit 5, and selects the compression encoding unit 71 according to the result of one selection.
Alternatively, the data regarding the rectangular area minus the rectangular area position information is sent to the compression encoding unit 72 . When the compression-encoded data is sent back from the compression-encoder 71 or the compression-encoder 72, the selection control unit 70 receives it and newly sets it as rectangular area compression-encoded data. The compression encoding method may be selected by a switch for each rectangular area, or the image signal within the rectangular area may be subjected to known feature extraction processing, such as mean value 1 variance value, histogram, frequency analysis, etc. The 0 selection control unit 70 may perform the 0 selection control unit 70, which compares the result with a threshold value and automatically selects a 0 according to the comparison result. Occur.

- In the above configuration and operation, claim (1) or (2)
) encoding of the image signal is realized.

The rectangular area compression encoded data, rectangular area position information, and compression method identification information generated by the compression encoding unit 7 in the previous operations are sent to the code storage unit 6.

The code storage unit 6 assigns a serial number to each rectangular area on a first-come, first-served basis, and uses this as an index to store rectangular area position information in the area position information storage format illustrated in FIG. As illustrated, rectangular area compression encoded data (1) is stored with compression method identification information (x) added thereto. When the code storage unit 6 completes the storage operation, it sends a completion signal to the arbitrary shape area setting unit 2. By performing the above operations for all the shape matrices stored in the arbitrary shape region setting section 2, claim (3) is achieved.
encoding of image signals is realized.

In the above explanation. An image signal output from an image input device such as a camera may be encoded as a moving image, or
Although it may be encoded as a still image, a moving image is composed of a plurality of temporally consecutive still images (frames).

It is easy to apply the encoding operation of the present invention for still images to each frame of a moving image.

Therefore, in view of the configuration of the present invention, either a moving image or a still image may be used.

(Effects of the Invention) Since the present invention is configured as described above, a rectangular area that includes an arbitrarily shaped area that is a closed area is set within the - images.

This is defined as a partial image according to claim (1) or (2).

Furthermore, as in claim (3), a plurality of partial images are provided within an image so as to cover all areas within the image, and each partial image is encoded and stored independently, thereby reproducing, for example, one image. At this time, only the arbitrarily shaped region within the image can be immediately extracted without reproducing the dummy signal portion or the portion outside the closed curve. In addition, the present invention allows partial images or a combination of several images to be used as a unit of search data in an image database, thereby reducing the need for partial display or processing of encoded pixels without complicated processing. This has the advantage that it can be achieved in less processing time.

4. Brief explanation of Figure cf+j Figure 1 is a block diagram of the configuration of the first embodiment of the present invention;
The figure shows an example of an image input from an image input device, FIG. 3 shows an example of a shape matrix, FIG. 4 shows an example of a rectangular area 71-RIX, and FIG. 5 shows a diagram explaining the position of the rectangular area in the image. , Fig. 6 is an example of the storage format of rectangular area position information, and Fig. 7 is an example of the storage format of rectangular area position information.
An example of the storage format of rectangular area encoded data in the embodiment, FIG. 8 is a block diagram of the configuration of the second embodiment of the present invention, and FIG. 9 is an example of the storage format of rectangular area encoded data in the second embodiment. FIG. 10 is a block diagram of the configuration of the data compression unit according to the third embodiment of the present invention, and FIG. 11 is an example of the storage format of rectangular area compressed encoded data in the third embodiment.

1... Image signal storage section, 2... Arbitrary shape area setting section, 3... Rectangular area setting section, 4... Dummy signal generation section, 5... Rectangular area encoding section, 51... - Intra-area encoding unit, 52... Shape encoding unit, 6... Code storage unit, 7... Data compression unit, 70... Selection control unit, 71.72... Compression encoding unit .

Patent Applicant: Nippon Telephone and Telegraph Co., Ltd. Figure 2 IU and 10I, 102, 103 tooth line + Il, 112, 113 Successive color shapes Figure 3 Figure 4 (2) Shadow area matrix) (a) (b )'
(c) Fig. 5 (yt type fertilization 4η toad information) 3g8 Fig. 9 Fig. 10 -------J Fig. 11

Claims (3)

[Claims] (1) When digitizing and encoding an image signal output from an image input device such as a camera, an arbitrary-shaped region that is a closed region and a rectangular region that includes the region are set in the image, and Among the pixels included in the rectangular area, for pixels in the arbitrary shape area, an image signal input from the image input device is encoded for the pixel, and for pixels not included in the arbitrary shape area, is an image signal encoding method characterized in that only a part of an input image is encoded by applying and encoding a dummy signal unrelated to the input image signal. (2) When digitizing and encoding an image signal output from an image input device such as a camera, an arbitrary-shaped region that is a closed region and a rectangular region that includes the region are set in the image, and a rectangular region that includes the region is set in the image. In a rectangular area, the input image is encoded by encoding information on closed curves that define the inner and outer boundaries of the arbitrary-shaped area and image signals input from the image input device for pixels within the arbitrary-shaped area. An image signal encoding method characterized by encoding only a portion of the image signal. (3) Setting rectangular areas at multiple locations in the image, and encoding information representing the position and size of the rectangular area within the image for each of the rectangular areas, and encoding the information over the entire input image. The image signal encoding method according to claim 1 or 2, characterized in that: