JP2000115785A - Coder - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide the coder that makes a task efficient so as to reduce a data quantity by providing a proper scale parameter to a coding means to attain the task of necessity of zero filling against a low efficiency in the case that the number of bits is less than the number of assigned bits due to a setting value of the scale parameter when a motion vector is coded. SOLUTION: A coding information detection means 303 decides a scale parameter 314 based on information 312 in a state of being photographed by a photographing means 302 and information 311 detected by a photographing information detection means 301 and a coding means 305 encodes a motion vector by using the parameter. A coding information decision means may decide the scale parameter based on the motion vector data by means of the coding means.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a camera-integrated VT
BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an encoding apparatus for securing a more suitable recording time by reducing the amount of generated codes in a system such as an R camera or a monitoring camera.

[0002]

2. Description of the Related Art Since an image signal has an enormous amount of data, it is inconvenient to record and transmit the image signal. Therefore, data is compressed using various compression techniques, and recording and transmission are performed. I have. Among these compression techniques, there is a motion vector as an element representing the relationship between the positions of the components of the image of the current frame and the image of the previous frame.

As shown in FIG. 5, a motion vector is obtained by calculating the average value of the pixels of a component 102 (referred to as a macroblock in MPEG) 102 of a current frame 101 and a search area (range) 104 in a target frame 103. Move sequentially to find the difference value of the average value of the pixels in the macroblock,
The motion vector is obtained by detecting the position where the difference value is minimum. Japanese Patent Laid-Open No. 8-140087 discloses a technique for efficiently detecting the motion vector.

[0004] When coding a motion vector obtained by the above-mentioned method, an image is immediately preceding a current macroblock by using the fact that the motion vector should be similar in an adjacent macroblock. Is encoded with the difference from the motion vector of the macroblock. The difference value (delta) is calculated based on the basic vector component (motion code), the scale parameter (f), and the residual vector component (motion resi
dual), and using Expression (1), delta = Sign (motion code) × [{Abs (motion code−1) × f} + motion residual + 1] (1) Here, the scale parameter f is allowed to be set for each picture, and is usually set in advance in accordance with the motion vector search range. That is, when the search range is fixed, a constant value is always set.

FIG. 6 shows a flow of processing for encoding a motion vector. In FIG. 6, reference numeral 212 denotes the absolute value of the difference value delta between the motion vector for the current macroblock and the motion vector for the immediately preceding macroblock. The motion code that satisfies Expression (1) and is the maximum is calculated with the value of 212. 213 is mo
is the maximum value of the action code, and 214 is the motion residua
l. Next, reference numeral 201 denotes a variable length encoding table for encoding a motion vector.
18-2 Annex B Corresponds to Table B-10. The variable length code 215 is obtained from the coded bits 211 of the motion vector difference value delta and the motion code 213 using the variable length coding table 201. This value is a value of 1 to 11 bits in MPEG2. On the other hand, motion residual 214 is f-1
It is output as bit data as it is as a remainder bit. For example, in the case of MPEG2, the scale parameter f = 6, and when delta = 33, the sign bit 21
1 is positive (0), motion code = 6, motion residual =
It becomes 2. At this time, the variable-length code data 202 becomes 00001000 in consideration of the code bits, and the remainder bit 203 becomes 00010.

[0006] These encodings are performed on a macroblock basis,
Further, since the processing is performed in the forward and backward directions and further in the horizontal and vertical directions, a maximum of eight motion vectors are required for one macroblock. If this is the case of MPEG MP {ML, 10800 motion vectors are required per picture.

[0007]

By the way, according to the above-described method of encoding a motion vector, the scale factor is determined depending on the search range. f-1 bits of code will be assigned. In other words, a code of f-1 bits is assigned to an image with a small motion (excess bits are padded with zeros), so that a meaningless code is generated.

On the other hand, image data has a huge data amount, and users have requested that the data amount be reduced to some extent for recording and transmission.

Accordingly, an object of the present invention is to provide an encoding device capable of increasing the encoding efficiency and reducing the data amount in such a configuration.

[0010]

According to a first aspect of the present invention, there is provided an encoding apparatus for setting a search area in a previous frame corresponding to a target image area in a current image frame based on an image captured by an image capturing means. Then, by sequentially moving within the search area, the absolute value of the difference between the image data of the image area of interest and the image data of the search area is obtained, and a motion vector is obtained from the search area in which the minimum value of the absolute difference is obtained. The difference value between the motion vector obtained in the same manner from another search area adjacent to the search area is represented by a basic vector component and a residual vector component using a scale parameter. An encoding apparatus configured to encode and assign the number of bits of a residual vector component by a scale parameter, and to obtain a motion of the imaging means itself. It includes an information detecting means, is characterized in that it has a shooting information determination means for determining a scale parameter by capturing information obtained from the imaging information detecting means.

According to the encoding apparatus of the first aspect, when encoding a motion vector, the scale parameter f is
It is determined according to the search range of the motion vector. That is, it is not changed normally unless the range of the motion vector changes, and this scale parameter is set by the user before starting the encoding. Therefore, the present invention predicts the range of the motion vector from the shooting state of the camera (camera, lens orientation, moving speed), sets a scale parameter f based on this information, and encodes the motion vector using this. By doing so, it is possible to improve the coding efficiency of the motion vector especially when there is little motion.

For this reason, in coding a motion vector, a scale parameter can be set efficiently, and useless data can be reduced. In particular, an advantageous effect can be obtained when a subject with a small motion is photographed from a video with a large motion.

According to a second aspect of the present invention, in the first aspect, the photographing information is angle information for adjusting an angle of a lens of the photographing means.

According to the coding apparatus of the second aspect, the same effect as that of the first aspect is obtained.

According to a third aspect of the present invention, in the encoding apparatus according to the first aspect, the photographing information is a maximum absolute value of motion information for a plurality of frames.

According to the encoding apparatus of the third aspect, the same effect as that of the first aspect is obtained.

According to a fourth aspect of the present invention, a search area is set in a previous frame corresponding to an image area of interest in a current image frame based on an image captured by an image capturing means. Are sequentially moved to obtain the absolute value of the difference between the image data of the image area of interest and the image data of the search area, obtain a motion vector from the search area in which the minimum value of the difference absolute value is obtained, and obtain a motion vector adjacent to the search area. A difference value from a motion vector similarly obtained from another search area is represented by a basic vector component and a residual vector component using a scale parameter, and the basic vector component and the residual component vector component are encoded, and the residual vector An encoding device configured such that the number of bits of the component is assigned by a scale parameter, wherein the scale parameter is a value of a previously encoded motion vector. It is characterized in that determined with reference to the relative value.

According to the encoding apparatus of the fourth aspect, the same effect as that of the first aspect is obtained.

According to a fifth aspect of the present invention, in the encoding apparatus according to the fourth aspect, the scale parameter is obtained by detecting motion information for a plurality of frames and using the maximum absolute value of the motion information.

According to the coding apparatus of the fifth aspect, the same effect as that of the fourth aspect is obtained.

[0021]

(Embodiment 1) An encoding apparatus according to a first embodiment of the present invention is shown in FIGS. In FIG. 1, reference numeral 301 denotes photographing information detecting means for detecting photographing information of the direction of a camera lens at the time of photographing. 3
Reference numeral 02 denotes a photographing unit for obtaining an image of a subject such as a camera, in which information such as focus and zoom that can be set by a photographer is also set. Reference numeral 303 denotes coded information determining means which is photographic information determining means for determining information necessary for coding using the information obtained at the time of shooting, and 304 temporarily stores video information obtained by the shooting means 302. 305 is a coding means for coding based on the data 314 and 315 obtained from the coding information determining means 303 and the memory means 304, respectively, and 306 is a coding means 305 It is a recording means for recording the code coded by. Reference numeral 311 denotes photographing information of the direction of the lens obtained by the photographing information detecting means 301, and reference numeral 312 denotes camera setting state information such as zoom and focus data at the time of photographing. These 311 and 31
The coding information determining means 303 determines a scale parameter necessary for coding the motion vector using the information of the second type. The method of determining the scale parameter will be described later. Reference numeral 313 denotes photographed video data, and this data is a target of image compression. These video data are temporarily stored in the memory means 304,
They are read out in the order in which they are encoded by the encoding means 305. 3
Reference numeral 16 denotes encoded bit stream data.

FIG. 2 is a diagram showing a configuration example of the photographing information detecting means 301 and the photographing means 302. Reference numeral 401 denotes an optical system which controls the angles of the vertical and horizontal photographing directions (lens directions) in accordance with a vertical control signal 422 and a horizontal control signal 424. Reference numeral 402 denotes a photoelectric conversion unit which converts a light signal 420 obtained by the optical system 401 into an electric signal.
D and the like. A signal processing unit 403 converts the electric signal 421 obtained by the photoelectric conversion unit 402 into a digital signal 411. A vertical motion detection sensor 405 for detecting vertical motion information of the photographing means 302
To detect the vertical direction photographing means movement signal 423, and input to the vertical direction lens control unit 404. A control signal 4 for controlling the vertical lens by the vertical lens controller 404
22 controls the optical system 401. Similarly, a horizontal direction motion detecting sensor 407 for detecting horizontal direction motion information of the image capturing unit 302 causes a horizontal direction photographing unit motion signal 42 to be detected.
5 is input to the horizontal lens control unit 406. The optical system 401 is controlled by the control signal 424 for controlling the horizontal lens by the horizontal lens control unit 406.

Here, the vertical direction photographing means movement signal 423 and the horizontal direction photographing means movement signal 425 which become angle information.
Is also output from the vertical and horizontal motion information output terminals 409 and 410 to the encoded information determining means 303.

FIG. 3 is a block diagram of the encoding information determining means 303, and includes the photographing means 302 and the photographing information detecting means 30.
Motion information for one to a plurality of frames is input, and a scale parameter is determined using the maximum absolute value of the motion information. That is, each of the output terminals 409 and 410 of the photographing information detecting means 301 inputs vertical and horizontal photographing means motion information to vertical and horizontal motion information input terminals 511 and 512, respectively. The circuit for selecting the maximum value of the absolute value is the motion information maximum value detection unit 5
01. The motion information maximum value detection unit 501 is configured to reset the maximum value of the absolute value by the reset signal 515 at intervals of one frame or more. From the maximum value motion information 513 detected by the motion information maximum value detection unit 501, in the scale parameter determination unit 502, dr is the focal length, θ is the angle in the horizontal and vertical directions, and θ is the angle for displaying the photographed object on the screen. Assuming that the ratio is α, the moving distance ds of the object on the screen is calculated from these parameters using Expression (2) (514),
This is reflected in the value of the scale parameter.

Ds = dr × tan θ × α (2) The scale parameter is determined by adding the effective bit width of the absolute value of ds to the initial scale parameter.

In the first embodiment, the moving distance d
Although the scale parameter is set according to the effective bit width of the absolute value of s, a certain value is set as a threshold, and when it exceeds the threshold, 1 is added to the current scale parameter,
A setting method such as a scale parameter is also possible.

Note that the encoded information determining means 303
Processing by a processor such as a microcomputer is also possible.

(Embodiment 2) An encoding apparatus according to a second embodiment of the present invention will be described with reference to FIG.

In FIG. 4, reference numeral 601 denotes a photographing means for photographing a subject, in which video data 611 obtained by the photographing means 601 is temporarily stored in a memory means 603, and is sequentially encoded by an encoding means 604. Is read (61
2). Here, the encoding unit 604 also performs a process of encoding a motion vector. Of the motion vectors detected at this time, the encoding unit 604 encodes motion vector data 613 encoded as a code, and determines encoding information. The coding information determination unit 602 also outputs to the scale parameter 61 based on the motion vector data 613.
4 is determined. The setting of the scale parameter is such that the picture unit is set to the shortest period and the group of pictures (MP
In the case of EG, this can be set as a GOP).

The coding information determining means 602 detects the maximum motion vector from the motion vector data 613 obtained from the coding means 604, and, based on the result, calculates the scale number based on the number of bits of the motion vector based on Table 1. Set parameters.

[0031]

[Table 1] Note that the encoded information determining means 602 can be processed by a processor such as a microcomputer as in the first embodiment.

[0032]

According to the encoding apparatus of the first aspect, when encoding a motion vector, the scale parameter f
Is determined according to the search range of the motion vector. That is, it is not changed normally unless the range of the motion vector changes, and this scale parameter is set by the user before starting the encoding. Therefore, the present invention predicts the range of the motion vector from the shooting state of the camera (camera, lens orientation, moving speed), sets a scale parameter f based on this information, and encodes the motion vector using this. By doing so, it is possible to improve the coding efficiency of the motion vector especially when there is little motion.

For this reason, in encoding a motion vector, a scale parameter can be set efficiently, and useless data can be reduced. In particular, an advantageous effect can be obtained when a subject with a small motion is photographed from a video with a large motion.

According to the encoding device of the second aspect, the same effect as that of the first aspect is obtained.

According to the encoding device of the third aspect, the same effect as that of the first aspect is obtained.

According to the encoding device of the fourth aspect, the same effect as that of the first aspect is obtained.

According to the coding apparatus of the fifth aspect, the same effect as that of the fourth aspect is obtained.

[Brief description of the drawings]

FIG. 1 is a block diagram of a first embodiment of the present invention.

FIG. 2 is a block diagram of a photographing information detecting means and a photographing means of the present invention.

FIG. 3 is a block diagram of an encoding information determining unit.

FIG. 4 is a block diagram of a second embodiment of the present invention.

FIG. 5 is a schematic explanatory diagram of a motion vector detection method.

FIG. 6 is an explanatory diagram showing encoding of a motion vector.

[Explanation of symbols]

 101 current frame 102 target area in current frame 103 target frame 104 target frame search area 105 target frame comparison area 301 shooting information detecting means 302 shooting means 303 coding information determining means 304 memory means 305 coding means 306 recording means 311 Vertical and horizontal shooting means motion information 312 shooting information (zoom, focus) 313 video signal 314 scale parameter 315 video signal (sorted) 316 bit stream 401 optical system 402 photoelectric conversion unit 403 signal processing unit 404 vertical lens control Unit 405 Vertical motion detection sensor 406 Horizontal lens control unit 407 Horizontal motion detection sensor 409 Vertical motion information output terminal 410 Horizontal motion information output terminal 411 Video signal output terminal 420 Video data (light) 421 Video signal 422 Vertical lens control signal 423 Vertical shooting means movement information 424 Horizontal lens control signal 425 Horizontal shooting means movement information 501 Motion information maximum value detection section 502 Scale parameter determination section 511 Vertical movement Information input terminal 512 Horizontal direction motion information input terminal 513 Motion information maximum value data 514 Scale parameter output terminal 601 Imaging means 602 Encoding information determining means 603 Memory means 604 Encoding means 605 Recording means 611 Video signal 612 Video signal (sorted) Thing) 613 motion vector 614 scale parameter 615 bit stream

 ──────────────────────────────────────────────────続 き Continued on the front page F term (reference) 5C018 FA02 FB03 5C053 GA10 GB19 GB30 GB37 KA03 KA24 LA01 5C059 KK19 NN03 NN11 NN19 SS14 TA63 TB07 TC12 TC42 TC47 TD02

Claims (5)

[Claims] 1. A search area is set in a previous frame corresponding to a target image area in a current image frame based on an image captured by an image capturing means, and the search area is sequentially moved in the search area. Taking the absolute value of the difference between the image data of the image area of interest and the image data of the search area, obtaining a motion vector from the search area in which the minimum value of the difference absolute value was obtained,
A difference value between a motion vector and a motion vector similarly obtained from another search area adjacent to the search area is represented by a basic vector component and a residual vector component using a scale parameter,
An encoding device configured to encode the basic vector component and the residual component vector component, and to assign the number of bits of the residual vector component by the scale parameter. An encoding apparatus comprising: a photographing information detecting unit; and a photographing information determining unit for determining the scale parameter based on photographing information obtained from the photographing information detecting unit. 2. The encoding apparatus according to claim 1, wherein the photographing information is angle information for adjusting an angle of a lens of the photographing means. 3. The encoding apparatus according to claim 1, wherein the shooting information is a maximum value of an absolute value among motion information for a plurality of frames. 4. A search area is set in a previous frame corresponding to a target image area in a current image frame based on an image captured by an image capturing means, and the search area is sequentially moved in the search area. Taking the absolute value of the difference between the image data of the image area of interest and the image data of the search area, obtaining a motion vector from the search area in which the minimum value of the difference absolute value was obtained,
A difference value between a motion vector and a motion vector similarly obtained from another search area adjacent to the search area is represented by a basic vector component and a residual vector component using a scale parameter,
An encoding apparatus configured to encode the basic vector component and the residual component vector component, wherein a number of bits of the residual vector component is assigned by the scale parameter. An encoding apparatus characterized in that it is determined with reference to an absolute value of a motion vector obtained. 5. The encoding device according to claim 4, wherein the scale parameter is determined by detecting motion information for a plurality of frames and using a maximum absolute value of the motion information.