JPH05197810A - Command input processing method by image - Google Patents

Abstract

(57) [Summary] [Object] An object of the present invention is to provide a simple command input means for inputting a command by the video itself, and particularly using a motion vector included in the image. [Structure] A command is extracted and input by detecting a motion vector and combining a spatial and temporal combination of the motion vector and a prepared dictionary representing a pattern of the motion vector.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention provides one means for inputting a command to a processing device, and in particular, a command is identified by a motion vector included in an image and the command is input. The present invention relates to a command input processing method using a captured image.

[0002]

2. Description of the Related Art FIG. 7 is a diagram for explaining the configuration of a general processing system. A command indicating a processing content is input to the processing device 2 by the input unit 1. The processing status (response) is displayed by the output means 3. In this system, the user can interactively proceed with the processing by using the input means 1 and the output means 3. In addition to the above, in a general processing apparatus, the data input 4 to be processed and the data output 5 to be the processing result may be separately connected to the processing apparatus.

Conventionally, a keyboard, a mouse, a touch panel, voice input, etc. have been generally used as command input means for this processing device.

[0004]

Although it is possible to use these methods as command input means for a video processing apparatus, in a system in which simple commands are input, these input means are structurally different from each other. Will be a burden. Further, when accessing the image processing apparatus from the user's standpoint, there is resistance in using a special input device.

That is, since the command input means and the processing data input means are separately provided, there is a problem that the interface becomes complicated. The present invention aims to realize a simple interface by inputting a command by the video itself, and particularly to provide a simple command input means using a motion vector included in an image.

[0006]

According to the present invention, a motion vector included in an image is detected, and a spatial and temporal combination of motion vectors and a dictionary representing a prepared motion vector pattern are used. I have to.

[0007]

The spatial and temporal combination of the above motion vectors is collated with the dictionary showing the motion vector pattern,
Extract the command to be input and input the command.

[0008]

DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 is an illustration of one embodiment of the present invention. In the figure, reference numeral 10 is a video input terminal, 11 is a motion vector detection circuit, 12 is a motion vector storage memory, 13 is a command determination circuit, 14 is a dictionary for command analysis, and 15 is a command output terminal.

The video signal input from the video input terminal 10 is input to the motion vector detection circuit 11. The motion vector detection circuit 11 divides an image into blocks composed of a plurality of pixels and detects a motion vector for each block.

The motion vector is a parameter value representing the amount of movement of the image from the immediately preceding video frame in each input video frame. In the field of normal image coding, 16 × 16 pixels are regarded as one block, and a two-dimensional vector that expresses where each block is most similar to the immediately preceding video frame within ± 15 pixels in the vertical and horizontal directions is used. It is used as a motion vector. That is, the motion vector detection circuit 11 outputs the motion vector V (i, j) of the block (i, j) every frame.

The motion vector V (i, j) is stored in the motion vector storage memory 12 for a period of several frames.
The command determination circuit 13 uses the vector value read from the motion vector storage memory 12 as a command analysis dictionary 14
The registered command is identified by matching with the pattern read from. The determined command is output from the command output terminal 15.

Details of the operation of the command determination circuit will be described with reference to FIGS. FIG. 2 shows that the motion vector of each block is quantized with a certain threshold and classified into states # 0 to # 8. Region # 0 indicates a region where the motion vector is small, that is, a region where there is no motion. # 1 to # 8 indicate areas in which motion vectors in the respective directions exist.

The motion vector of each block is represented in this way, and the result is temporally accumulated. FIG. 3 shows this state. Vector values # 0 to # 8 are sequentially input from each block for each frame and stored in the memory over several frames. The example of FIG. 3 shows that the vector in the # 5 direction is generated after the vector in the # 1 direction is generated.

(C1) to (C4) shown in FIG. 4 are examples of patterns stored in the dictionary for command analysis, and particularly show patterns relating to temporal order relation of motion vectors.

From the result of FIG. 3, if only the order relation in several frames is extracted, the result of (C1) in FIG. 4 is obtained (C1).
The command corresponding to 1) is detected. (C1) ~ (C
4) corresponds to the case where the motion vector is rotating up and down, left and right, rotating left, and rotating right, respectively.

(D1) to (D8) shown in FIG. 5 are examples of patterns similarly stored in the dictionary for command analysis, and particularly show spatial arrangement patterns of motion vectors. When different motion vectors are detected between adjacent or nearby blocks, the commands corresponding to (D1) to (D8) are detected.

The above example of the motion vector pattern is assumed to be generated by the human image present in the image. For example, a command (C1) is issued by moving the head image up and down.
The command (C2) is detected by the horizontal movement. By expanding and closing both hands, the command (D
1) and (D2) are detected. Similarly, a command system can be constructed by registering a command corresponding to an effective motion vector pattern.

In the above operation, the performance is determined by how correctly the general motion (motion not for inputting a command) included in the image and the motion for command input are separated and the command can be input correctly. .. Therefore, although the temporal combination, the spatial combination, and the classification / quantization based on the size of the vector are applied as the motion pattern, a temporal / spatial window can be further provided. That is, by detecting the command only in a certain area in the screen, it is possible to avoid the influence from other than the command area. In addition, limiting the time period for command detection to only the period that requires command input has the same effect. When inputting a quantitative instruction amount as a command, the size of the motion vector, the interval between blocks in which the motion vector has occurred, the duration during which the motion vector has occurred, the number of times the motion vector has changed, etc. It can be used in combination with.

The mode of use of the present invention will be specifically described with reference to FIG. 10 is a video input terminal, 11 is a motion vector detection circuit, 12 is a motion vector storage memory, 13 is a command determination circuit, 14 is a command analysis dictionary, 15 is a command output terminal, 16 is an encoding circuit, and 17 is a video storage. A memory, 18 is a transmission device, and 19 is a line connection terminal.

The encoding circuit 16 encodes the video signal from the video input terminal and outputs it to the transmission device 18. Transmission device 18
Outputs encoded data from the line connection terminal 19. The coding circuit 16 uses the output of the motion vector detection circuit 11 to perform motion compensation predictive coding. That is, one of the gist of the present invention is to realize the command input with good man-machine interface in a simplified manner by using the motion vector detection function originally possessed by the encoding processing unit as the command input means. Especially. In order to specify the processing to be executed by the command, for example, the usage condition of the video storage memory 17 is specified. For example, the (D1) command causes the video storage memory 17 to execute a process for starting the storage of the input video, and the (D2) command ends the process. Also, the data in the video storage memory 17 is transmitted by the (C1) command, and is ended by the (C2) command.

Needless to say, the present invention can be applied to a decoding circuit. That is, the transmitting unit transmits a video containing a command, and the receiving unit performs command detection using the decoded motion vector.

[0022]

As described above, according to the present invention, a command can be input by using an input video, so that it can be used as a simple input means of a video processing apparatus. Further, since no special input means is required, a good man-machine interface can be realized.

[Brief description of drawings]

FIG. 1 is an explanatory diagram of an embodiment of the present invention.

FIG. 2 is an operation explanatory diagram of a command determination circuit.

FIG. 3 is an operation explanatory diagram of a command determination circuit.

FIG. 4 is an operation explanatory diagram of a command determination circuit.

FIG. 5 is an operation explanatory diagram of a command determination circuit.

FIG. 6 is a diagram illustrating a usage mode of the present invention.

FIG. 7 is an explanatory diagram of a general processing system.

[Explanation of symbols]

 10 Video Input Terminal 11 Motion Vector Detection Circuit 12 Motion Vector Storage Memory 13 Command Judgment Circuit 14 Dictionary for Command Analysis 15 Command Output Terminal

Claims (1)

[Claims] 1. A motion vector included in an image is detected,
In a command input processing method for inputting a command corresponding to the motion vector to a data processing device, a spatial and temporal combination of the above motion vectors and a dictionary representing a motion vector pattern prepared in advance are used to display an input video image. A command input processing method using video, characterized in that a command corresponding to a contained motion vector is identified.