CN100336375C - Portable terminal device and character input method - Google Patents
Portable terminal device and character input method Download PDFInfo
- Publication number
- CN100336375C CN100336375C CNB2004100333701A CN200410033370A CN100336375C CN 100336375 C CN100336375 C CN 100336375C CN B2004100333701 A CNB2004100333701 A CN B2004100333701A CN 200410033370 A CN200410033370 A CN 200410033370A CN 100336375 C CN100336375 C CN 100336375C
- Authority
- CN
- China
- Prior art keywords
- motion
- vector
- input
- literal
- terminal device
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Images
Abstract
To provide a portable terminal device that can improve character input operation performance by enabling a character input with minimum keystrokes.The portable terminal device 1 comprises an imaging part 3 for picking up an image, a motion vector detection part 4 for detecting a motion vector from an image signal outputted from the imaging part 3, a key operation part 5 having a plurality of keys, a character input decision part 6, and a display part 7. The character input decision part 6 selects, converts and decides input characters according to outputs from the key operation part 5 and outputs from the motion vector detection part 4. The display part 7 displays characters selected, converted and decided by the character input decision part 6. (C)2005,JPO&NCIPI.
Description
Technical field
The mobile phone that the present invention relates to have the literal input function that is used to generate telephone directory or Email is the mobile terminal device of representative.
Background technology
In the past, in mobile terminal devices such as mobile phone,, generally used the function key of the numerical key of keypad and multifunction key etc., carried out the literal input by following operation as the literal input block that is used to generate telephone directory and Email etc.At first, trigger input operation by numerical key, the literal phonetic that input should be imported, then pass through the map function of function of use key, show the corresponding a plurality of Chinese character conversion candidates of phonetic that finish with described input, by definite operation of function of use key, select and definite required Chinese character at last from described Chinese character conversion candidate.Promptly in described literal input block, the literal input operation all depends on key operation.
As concrete example, for example in general mobile phone, mode as described phonetic input, employing distributes alphabet respectively to " 2 "～" 9 " key that constitutes numerical key, if select to press key arbitrarily in the middle of them, according to the number of times of pressing key, change the alphabetic(al) mode of distributing to this key that shows successively.
Situation when for example, considering input " hello ".At this moment, if press " 6 " key, " n " of input alphabet twice.In addition, if press " 4 " key three times, with regard to " i " key of input alphabet.By the map function of function of use key, show the corresponding a plurality of Chinese character conversion candidates of phonetic " ni " that finish with described input, by definite operation of function of use key, from described Chinese character conversion candidate, select to determine required Chinese character " you " at last.Then, if press " 4 " key twice, with regard to " h " of input alphabet.In addition, if press " 2 " key next time, with regard to " a " of input alphabet.If press " 6 " key again three times, with regard to " o " of input alphabet.Then,, show the corresponding a plurality of Chinese character conversion candidates of phonetic " hao " that finish with described input, by definite operation of function key, from described Chinese character conversion candidate, select to determine required Chinese character " good " at last by definite operation of function of use key.
, in such literal input operation, in order to import target text, numerical key press increased frequency, have the problem of operability difference.
Therefore,, propose when the input Japanese as an example of eliminating the portable terminal of such operability difference in the example in the past, 50 sounds are configured to rectangular, the touch potential (for example patent documentation 1) the during input operation of minimizing literal.
In addition, as other examples, the acceleration detection parts that use acceleration transducer have also been proposed to be provided with, detection output according to these acceleration detection parts, obtain the motion track that moves that acts on the mobile terminal device, on display, show this motion track, if under the state that shows this motion track, determine operation, just the mobile terminal device (for example patent documentation 2) of described motion track as input characters identification.At this moment, can be as the handwriting input input characters.In addition, in this patent documentation 2, propose not rely on the operation of function key, carry out the selection or the affirmation of input characters by detecting mobile direction and amount of movement.
[patent documentation 1]
The spy opens the 2000-172417 communique
[patent documentation 2]
The spy opens the 2002-169645 communique
, in the mobile terminal device of above-mentioned patent documentation 1, for literal input, the key that 1 literal subsistence level is twice is pressed.In addition, if press by primary key, the corresponding lists that shows the literal that next key is pressed and import in view of the above, to improve operability, then need as primary input operation → demonstration → secondary input operation, add between the input of front and back is handled and show and handle, make to handle to become complicated, existence can't adapt to the problem that the high speed of input operation requires.
In addition, in the mobile terminal device of patent documentation 2, be necessary to be attached to acceleration transducer unnecessary in the original mobile terminal device, the function that detects moving direction and amount of movement behind 2 integrations of its output is also necessitated, for with this function of done with high accuracy, there is the problem complicated, high price that becomes of handling.
Summary of the invention
The present invention is in view of the above problems and proposes that its purpose is: provide to press with minimal key and can realize that reliable literal imports the mobile terminal device of raising literal input operation.
Mobile terminal device of the present invention comprises: the image pickup part of taking image; Detect the motion-vector test section of the motion-vector of whole image according to picture intelligence from described image pickup part output; Key operation portion with a plurality of keys; According to from the output of described key operation portion with from the output of described motion-vector test section, selection, conversion and definite inputting words input determination portion; With the display part that shows by the selection of described literal input determination portion, conversion and definite literal.
According to mobile terminal device of the present invention, owing to have the image pickup part of taking image; Detect the motion-vector test section of motion-vector according to picture intelligence by described image pickup part output; Key operation portion with a plurality of keys; According to from the output of described key operation portion, from the output of described motion-vector test section, selection, conversion and definite inputting words input determination portion; Demonstration is by the display part of the selection of described literal input determination portion, conversion and definite literal, therefore, can detect the gesture actions that mobile terminal device is shaken about up and down with the motion-vector checkout gear after, be applied in the literal input information.
Character input method of the present invention comprises: the shooting step of taking image; The motion-vector that detects the motion-vector of whole image according to the picture intelligence from described shooting step output detects step; The step of carrying out input operation from key operation portion with a plurality of keys; According to output and described motion-vector from described key operation step, selection, conversion and definite inputting words input determining step; With the step display that shows by the selection of described literal input determining step, conversion and definite literal.
Description of drawings
Following brief description accompanying drawing.
Fig. 1 is the block diagram of schematic configuration of literal input mechanism of the mobile terminal device of the expression embodiment of the invention 1.
Fig. 2 is the key diagram of the posture input action of this mobile terminal device of expression.
Fig. 3 is the block diagram of circuit one example of the expression motion-vector detection part that constitutes this mobile terminal device.
Fig. 4 is the block diagram of other circuit examples of this motion-vector detection part of expression.
Fig. 5 is the figure that the literal matrix of map table 1 pattern of storing in the storage part of determination portion imported in expression mobile terminal device literal.
Fig. 6 is the figure that the literal matrix of map table 2 patterns of storing in the storage part of determination portion imported in expression mobile terminal device literal.
Fig. 7 is the figure that the motion-vector time series of the mobile terminal device detection of the expression embodiment of the invention 2 changes.
Fig. 8 is the figure of handling process of the motion-vector detection method of this mobile terminal device of expression.
Fig. 9 is the figure of cutting state of frame of the mobile terminal device of the expression embodiment of the invention 3.
Figure 10 is the figure of the motion-vector detected state of this mobile terminal device of expression.
Figure 11 is the figure of handling process of the motion-vector detection method of this mobile terminal device of expression.
Embodiment
Fig. 1 is the block diagram of schematic configuration of literal input mechanism of the mobile terminal device of the expression embodiment of the invention, and Fig. 2 is the stereoscopic figure of this mobile terminal device.This mobile terminal device 1 (Fig. 2) is the mobile phone of band camera function, and literal input mechanism 2 (Fig. 1) has image pickup part 3, motion-vector test section 4, key operation portion 5, literal input determination portion 6 and display part 7.
Motion-vector test section 4 is to export the parts that detect motion-vector according to the image of image pickup part 3, and particularly, the relativeness of the image of interframe is calculated and moved to which direction at front and back interframe image before and after utilizing.In this motion-vector test section 4, a part is also used as the addressable part that is used for the animation compression when sending animation and the motion-vector testing circuit (detecting the motion-vector of each macro block unit of picture) prepared.
Here, use block diagram shown in Figure 3, the structure example of motion-vector test section 4 is described.
Carry out the A/D conversion from the picture intelligence of image pickup part 3 outputs by A/D converter 41, these signals store in the preceding frame block storage 42 by each macro block unit of picture.Parallel therewith, pixel walking circuit 43 is taken into the picture intelligence that pixel location (address) from each macro block moves to the pixel of moving direction each position different with amount of movement, these signals is input in the absolute value summing circuit 44 of next stage.In absolute value summing circuit 44, by each macro block unit, obtain the former frame that frame block storage 42 in the past reads each macro block the pixel absolute value and make the summation of pixel absolute value of each macro block of the next frame that pixel location moves with pixel shifter 43.
In each piece minimum detection circuit 45 of next stage, after the minimum value of the absolute value summation of obtaining each macro block, detect motion-vector.Then, in the averaging circuit 46 of next stage, by in picture the motion-vector value equalization of each macro block, and as the motion-vector of whole image.
In motion-vector test section 4, the part from A/D converter 41 to absolute value summing circuit 44 is as the addressable part that is used for animation compression and holds a concurrent post the circuit part of the circuit that mobile phone had that drives the picture function.In view of the above, can simplify the structure of motion-vector test section 4, can reduce cost.
In addition, use block diagram shown in Figure 4, other structure example of motion-vector test section 4 are described.It should be noted that more specific description is carried out in No. the 2641599th, patent, so be only limited to simple declaration following.
In the present example, by the representative point matching method that in manual correction, uses, detect the motion-vector of whole image.Be that the representative point matching method is to set a plurality of representative points in picture, the signal value of the representative point of storage former frame, and obtain absolute value with the difference of the signal value of the representative point that is positioned at present frame pixel on every side, according to this absolute value, detect the method for motion-vector.
More specifically, as shown in Figure 4, carry out the A/D conversion from the picture intelligence of image pickup part 3 output by A/D converter 41 after, frame representative point memory 42A and absolute value circuit 44A before offering.
The signal value of the representative point of the former frame in preceding frame representative point memory 42A before storage 1 frame, use absolute value circuit 44A, obtain absolute value of each difference of the representative point of each pixel that is positioned at the address space identical distance and former frame according to the representative point that behind 1 frame is present frame with the table 47 that adds up.
Then, on the corresponding address of table 47 that adds up, the add up some fractional part of representative point of this absolute value, in the table comparison circuit 48 of next stage, calculate lowest address with accumulated value, in 1 frame position of image how much to which direction moved be the motion-vector value.
By adopting such structure, the higher motion-vector of precision is detected becomes possibility.In addition, when having the manual correction function in the mobile phone that drives the picture function, can simplify the structure of motion-vector test section 4, reduce cost by the part of dual-purpose circuit.
Here, get back to the explanation of Fig. 2.Key operation portion 5 is the input blocks that use in the input such as literal, is made of a plurality of keys, has " 1 "～" 0 " key and other function keys that constitute numerical key, behind the key input coding that the user is pressed, is input to literal input determination portion 6.
Literal input determination portion 6 is according to the expectant control program, and according to from the output of key operation portion 5 with from the output of motion-vector test section 4, selection, conversion and definite parts of the literal of control input are made of control part 61 and storage part 62.Control part 61 is according to the input from key operation portion 5 and motion-vector test section 4, carries out the selection, conversion of literal input pattern, each definite step.In storage part 62,, store the pre-prepd table information (key being imported and detected the matrix that motion-vector and literal are associated) that literal is selected that is used for according to a plurality of literal input patterns.
Below, the literal input operation of mobile terminal device 1 is described with reference to Fig. 2.
At first, make image pickup part 3 work, become shooting and be taken arbitrarily after the pattern of body M, obtain picture intelligence.It should be noted that, in Fig. 2, show the body M that is taken that takes at display part 7, but its purpose is to show literal or its map function of input, so need not show the body M that is taken.This body M that is taken can be arbitrarily basically, has the independent specific body that is taken that moves respectively with personage etc. and compares, and more wishes to move the landscape of almost fixing.In addition, because the dependency relation of the image of interframe before and after in motion-vector detects, utilizing, so can not carry out the adjustment of focal length etc.
Here, in Fig. 2, arrow a～d represents to act on the physics moving direction of mobile terminal device 1, and image representation mobile terminal device 1 (A)～(D) moves (motion-vector) from the image of the body M that is taken that is taken by image pickup part 3 of inactive state when arrow a～d tilts like that.Arrow a represents that the top of mobile terminal device 1 is to the back side one survey inclination, arrow b represents that the right side of mobile terminal device 1 is to the back side one survey inclination, arrow c represents the bottom of mobile terminal device 1 to the back side one survey inclination, and arrow d represents that the left side of mobile terminal device 1 is to the back side one survey inclination.
With mobile corresponding (A) image of arrow a in, move (motion-vector) of image becomes upwards, with mobile corresponding (B) image of arrow b in, motion-vector becomes to the right, with mobile corresponding (C) image of arrow c in, motion-vector becomes downwards, with mobile corresponding (D) image of arrow d in, motion-vector becomes left.Motion-vector test section 4 detects these motion-vectors.
Therefore, when with the motion-vector value of motion-vector test section 4 detected left and right directions and above-below direction during at preset range, for example motion-vector to the right is bigger than predetermined value, the motion-vector of above-below direction is than predetermined value hour, and the literal input determination portion 6 that receives the output of motion-vector test section 4 judges that users have carried out the input of arrow b posture to mobile terminal device 1.
Equally, when at left and right directions and above-below direction, all than predetermined value hour, literal input determination portion 6 judges that mobile terminal devices 1 are " static " to motion-vector.If is the situation of this " static " situation of (O) as motion-vector, by (A), (B), (C), (D) are added this motion-vector (O), can suppose 5 kinds of posture inputs, export these 5 kinds of information to literal input determination portion 6 from motion-vector test section 4.
Below, with reference to Fig. 5, Fig. 6, concrete character input method is described.
Fig. 5, Fig. 6 are the concrete examples of the expression table information corresponding with the literal input pattern.
Matrix table shown in Figure 5 is the example that constitutes according to the alphabet input of using in the universal telephone in the world.For example " J ", " K ", " L " distribute to " 5 " key, thus these three literal distribute to the key input of " 5 " key and (A), the posture input of (B), (C).This literal input pattern is called " map table 1 " pattern.Matrix table shown in Figure 6 is the example that the alphabet input of " 2 touch input " of use in importing according to the literal of the pocket beeper of Japan constitutes.The i.e. input that touches for the second time is transformed to moving of (O), (A), (B), (C), (D), and this literal input pattern is called " map table 2 " pattern.It should be noted that according to carry out the definite of literal input pattern from the input of key operation portion 5, this input pattern shows on display part 7.
In literal input determination portion 6, at first as the input pattern switch step of first step, control part 61 is determined the literal input pattern according to the input from key operation portion 5, shows its input pattern on display part 7.Select in the step at the input characters of conduct second step of following, the table information (matrix) that control part 61 is read this input pattern from storage part 62 is selected literal according to this matrix.Then, in the shift step of following as third step, control part 61 is the text transform of selecting a Chinese character, in the definite step as the 4th step, selects, determines required Chinese character from conversion Chinese character candidate.
Like this, by the processing of literal input determination portion 6 action is divided into the multistage, by carrying out 1 key input and 1 posture input, or carry out the two simultaneously, can select the literal imported, can further improve the operability that literal is imported.
Below, illustrate as situation concrete example, when importing Chinese character " hello ".At first, the function key with key operation portion 5 is set at the literal input pattern " map table 1 ".In " map table 1 " pattern, read the matrix of Fig. 5 from storage part 62 after, become the input holding state.Under this state, if according to the matrix of Fig. 5, select " n " with " 6 " key and the input of posture (B), select " i " with " 4 " key and the input of posture (C), just demonstration " ni " on display part 7.In this case, from the output of key operation portion 5 and independently of one another, so can import simultaneously from the output of motion-vector test section 4.It is 2 times that the key that uses in the input of this " ni " is pressed number of times.
Then, carry out map function with function key (general " to Down Arrow ").Like this, because except " you ", " you " the undesired Chinese characters such as Chinese character in addition with " ni " expression enter candidate sometimes, so for the Chinese character of selecting to determine to want, determine operation (general, as, to determine) with function key with " determining key " with " up-down arrow " selection.
If select " h " with " 4 " key and posture input (B), select " a " with " 2 " key and posture input (A), select " o " with " 6 " key and posture input (C), just on display part 7, show " hao ".It is 3 times that the key that uses in the input of this " hao " is pressed number of times.
Then, carry out map function with function key.Like this, " good " in addition, " good " the undesired Chinese characters such as Chinese character in addition with " hao " expression enter candidate sometimes, so the Chinese character in order to select to determine to want is determined to operate with function key.
Below, the situation when input characters " Hello " is described.At this moment, at first be set at " map table 1 " pattern, matrix according to Fig. 5, select " H " with " 4 " key and the input of posture (B), change to the small letter input pattern with " 8 " key and the input of posture (D), select " e ", select " l " with " 5 " key and the input of posture (C) with " 3 " key and the input of posture (B), select " l " with " 5 " key and the input of posture (C), select " o " with " 6 " key and the input of posture (C).At this moment, just can import " Hello " by 6 buttons.
It should be noted that, when map table 2 patterns, according to the matrix input of Fig. 6.At this moment, can switch capitalization input and small letter input by the function of use bond energy.
If conclude above action, then can be by changing the angle of mobile terminal device 1 with respect to the body that is taken, change the camera angle of the body 1 that is taken, the variation of camera angle is considered as the mobile of body M that be taken, detect motion-vector, can be that posture replaces key operation with respect to the angle of the body that is taken with changing mobile terminal device 1, can carry out reliable literal input, can improve the operability of literal input with minimal key operation number of times.
It should be noted that in the above description, posture is input as 5 kinds, but this is the result who simplifies in order to reduce flase drop to survey, except static, about, up and down, can comprise moving of incline direction etc. wherein, to increase the kind of posture input as a kind of of posture input.
In addition, in the above description, the situation of carrying out the operation of key operation and posture simultaneously is described, but also can for example behind key operation, carries out the posture operation not simultaneously.Which at this moment, also consider to append: behind key operation, on display part 7, show with posture operation then to select literal; Or set when being used for not having posture to import during certain, it is judged to be the time-out time of (O) posture.
In addition, in mobile terminal device shown in Figure 21, the back side one side that is illustrated in the operating surface 1a that key operation portion 5 and display part 7 are set is provided with the example of image pickup part 3, but also can be arranged on operating surface 1a one side to image pickup part 3.At this moment, can even mobile terminal device itself is motionless, also can carry out the literal input operation operator self as the body M that is taken with body actions such as shaking the head of operator.
Below, in this enforcement side, illustrate that the seasonal effect in time series that detects motion-vector changes after, improve the method for the accuracy of detection of motion-vector.
As described in example 1 above,, compare, more wish to have only the scene of the fixedly landscape that does not almost move with the scene that has mobile objects such as the personage that is moved respectively or animal as the body M that is taken that takes with image pickup part 3.Its reason is: if mobile object is present in the image pickup scope, detect the motion-vector corresponding with this mobile object sometimes.As a result, the flase drop of motion-vector is surveyed to be increased, and accuracy of detection descends., even select to have only the body that is taken of landscape, because human or animal etc. cross image pickup scope, the flase drop of the vector that is moved is sometimes surveyed.
Fig. 7 is the figure that changes the time of expression motion-vector, and the longitudinal axis is a motion-vector, and transverse axis is the time.For example when selecting landscape when being taken body, perhaps when mobile terminal device 1 self is static, suppose the situation that mobile object such as people's grade crosses the image pickup scope of image pickup part 3.At this moment, the detected state of motion-vector is shown in Fig. 7 (A), and at (a) regularly, mobile object enters the image pickup scope of image pickup part 3, leaves image pickup scope constantly at (b).When crossing before the landscape that mobile object is not almost having to move like this, detected motion-vector becomes and has only a direction.
And Fig. 7 (B) expression selects landscape as being taken body, for the user carries out the literal input operation, makes mobile terminal device 1 resemble among Fig. 2 the detected state of the motion-vector when mobile arrow a～d.After the user for example resembled mobile terminal device 1 to move the arrow b, the display part 7 for the state of confirming display text input or conversion action made mobile terminal device 1 move to the direction opposite with arrow b again.Promptly regularly, begin to make portable terminal 1 to move, finish this action constantly at (d) to the direction of arrow b at (c).Then, portable terminal 1 beginning is moved to the direction opposite with arrow b, at (f) constantly, finish this action at (e).Like this, when carrying out the literal input, at first detect the motion-vector of certain direction, then, detect opposite motion-vector.
Therefore, handle, can differentiate flase drop survey that interference causes and user's literal input operation by the detection of carrying out motion-vector with flow process shown in Figure 8.
At first, detect the motion-vector of scheduled volume in step 1.Then in step 2, set the timer of scheduled period, during timer in, judge whether detect and the reverse motion-vector (step S3, S4) of motion-vector that in step 1, detects.When detecting reverse motion-vector, be judged as user's literal input operation (step S5).
By such after detecting a motion-vector on the direction, the situation that detects reverse motion-vector in the scheduled period is judged as user's literal input operation, can reduce the flase drop of motion-vector and survey, and can improve the precision of detection.
In handling process shown in Figure 8, after the motion-vector of step 1 detects, in step 2, set timer., with the input of posture that mobile terminal device 1 is moved simultaneously, or before the posture input, have input, from the input of the numerical key corresponding with matrix shown in Figure 5 from the function key that is used for determining the literal input pattern.Therefore, also can carry out the setting of timer these key inputs as triggering.
The storage part 62 that the cell-phone terminal device of present embodiment for example makes literal shown in Figure 1 import determination portion 6 has the function at the detected motion-vector of scheduled period stored.Promptly according to handling process shown in Figure 8, the motion-vector that use is detected by motion-vector test section 4, be stored in the motion-vector of table information in the storage part 62, storage, from the input of key operation portion 5, by control part 62 carry out literal input pattern, selection, conversion, each step such as determine.It should be noted that, except storage part 62, also can be provided for storage part (not shown) at the detected motion-vector of scheduled period stored.
In the present embodiment, illustrate frame is divided into a plurality of zones,, improve the method for the accuracy of detection of motion-vector by using the frequency distribution of the detected motion-vector in ground, region-by-region.
At first, frame is divided into a plurality of zones, Fig. 9 (A) represents entire frame, and Fig. 9 (B) expression is divided into for example state of 16 parts to the frame of Fig. 9 (A).It should be noted that the number of cutting apart of frame also can be beyond 16, the regional shape of each that cut apart also can be the shape beyond the rectangle in addition.
Then, cut zone ground detects motion-vector one by one.When Figure 10 (A) is illustrated in the posture input, the detected state of the motion-vector when mobile object crosses image pickup scope.In addition, during the input of Figure 10 (B) expression posture, the detected state of the motion-vector when not having mobile object in the image pickup scope.In Figure 10 (A), mobile object occupies 9 upper left zones, and other 7 remaining zones are backgrounds.It is the ratio background height that mobile object accounts for entire frame.Therefore, at this moment detect in the method for peak-peak, detect the motion-vector corresponding with mobile object in frequency distribution according to motion-vector.
, when the situation shown in Figure 10 (A), detect two motion-vectors of existence in this frame, these two motion-vectors are as the candidate of two motion-vectors representing this frame.And, if be present in frame before or after the frame shown in Figure 10 (A) in time if the frame shown in Figure 10 (B), then by relatively both, the motion-vector that can detect the frame of representing Figure 10 (A) is a motion-vector to the right.Promptly in each frame, detect two peak values of the frequency distribution of motion-vector,, judge which is the motion-vector of this frame of representative in two motion-vectors by comparing with the frame that is positioned at front and back in time.By this processing, even there is mobile object etc., also can reduces the flase drop of motion-vector and survey, improve accuracy of detection.
Figure 11 represents the handling process of detection method of the motion-vector of present embodiment.At first, in step 11, detect motion-vector in each zone after frame is cut apart.Then use the motion-vector that in each zone, detects to detect the frequency distribution (step 12) of motion-vector.Whether the frequency distribution peak value that detects motion-vector is 1, and (step 13) when being one, judges that detected peak value is a motion-vector (step 14) of representing this frame.At this moment, diminish owing to move the confusion that causes, the peak value that can judge detection is the motion-vector corresponding with background.
And in step 13, when the frequency distribution peak value of judging motion-vector was not one, whether detection peak was two (step S15).When peak value is two, with liken in time for the frame of process object forward or lean on after frame relatively, judge which peak value is the motion-vector (step 16) of representative frame.When there are two in peak value, can consider that one is the motion-vector corresponding with background, another is the motion-vector corresponding with mobile object.Therefore, both during as candidate, by be positioned in time before and after frame relatively, carry out the judgement of motion-vector, even for for the frame of process object, the ratio that background accounts for is less, also can detect the motion-vector corresponding with background.Handle by this, the flase drop of motion-vector is surveyed and is reduced, and accuracy of detection improves.
, when the frequency distribution peak value that detects motion-vector in step 15 is 2 when above, the image that is judged as in the frame is disorderly, is judged to be the motion-vector (step 16) that does not detect representative frame.
It should be noted that, in step 16, illustrated and used the picture that is positioned in time as the front and back of the present frame of process object as the frame that is used for comparison, but also can be only with forward in time former frame relatively, in addition, also can be only with lean in time after a back frame relatively.Frame relatively is not limited to as the former frame of the present frame of process object or back one frame, also can be frame before the scheduled time or frame afterwards, or the mean value of the frame of predetermined number.In addition, the motion-vector that can compare representative frame each other, the motion-vector that also comprises of candidate each other or the frequency distribution of motion-vector self.
In addition, the peak value that detects from the frequency distribution of motion-vector is 2, but can be more than 2.
The cell-phone terminal device of present embodiment for example makes the storage part 62 of literal input determination portion 6 shown in Figure 1 have the function of the detected motion-vector of storage in the scheduled period.Promptly according to handling process shown in Figure 11, the motion-vector that use is detected by motion-vector test section 4, be stored in the motion-vector of table information in the storage part 62, storage, from the input of key operation portion 5, by control part 62 carry out literal input pattern, selection, conversion, determine each step.It should be noted that, except storage part 62, also can be provided for storage part (not shown) at the detected motion-vector of scheduled period stored.
It should be noted that, in this manual, be that example is carried out explanation of the present invention with Chinese, but mobile phone of the present invention and character input method are not only applicable to Chinese, can be applicable to other language yet.
Claims (8)
1. mobile terminal device comprises:
Take the image pickup part of image;
Detect the motion-vector test section of the motion-vector of whole image according to picture intelligence from described image pickup part output;
Key operation portion with a plurality of keys;
According to from the output of described key operation portion with from the output of described motion-vector test section, selection, conversion and definite inputting words input determination portion; With
Demonstration is by the display part of the selection of described literal input determination portion, conversion and definite literal.
2. mobile terminal device according to claim 1, wherein:
Described motion-vector test section has the motion-vector testing circuit that uses in the compression of described picture intelligence.
3. mobile terminal device according to claim 1, wherein:
Described motion-vector test section has: the motion-vector testing circuit, be used in picture, setting a plurality of representative points, the signal value of the described representative point of frame before the storage, and obtain and be positioned at the absolute value of difference of the signal value of the pixel around the described representative point of present frame, utilization detects the motion-vector of whole image according to the representative point matching method of this absolute value detection motion-vector.
4. mobile terminal device according to claim 1, wherein:
Described literal input determination portion has control part and storage part;
Described control part is carried out following steps:
From predefined kinds of words input pattern, select a kind of first step of pattern;
Read the Word message of the pattern of from the described storage part of storage and described each literal input pattern word information relates, selecting, according to predefined accordingly described key operation of described Word message and detected motion-vector, select second step of corresponding input characters;
The third step of the selected literal of conversion; With
Determine the 4th step of the literal of institute's conversion.
5. mobile terminal device according to claim 4, wherein:
Described second step comprises: by changing described mobile terminal device, and change the camera angle of the described body that is taken with respect to the angle of the body that is taken, the variation of camera angle is considered as moving of the described body that is taken after, detect the step of described motion-vector.
6. character input method comprises:
Take the shooting step of image;
The motion-vector that detects the motion-vector of whole image according to the picture intelligence from described shooting step output detects step;
The step of carrying out input operation from key operation portion with a plurality of keys;
According to output and described motion-vector from described key operation step, selection, conversion and definite inputting words input determining step; With
Demonstration is by the step display of the selection of described literal input determining step, conversion and definite literal.
7. character input method according to claim 6 also comprises:
Detect the time series change-detection step of the time series variation of motion-vector;
Wherein literal input determining step is according to from the output of this time series change-detection step with from the output of key operation step, selection, conversion and definite input characters.
8. according to claim 6 or 7 described character input methods, also comprise:
Frame is divided into a plurality of zones, and the frequency distribution that detects the frequency distribution of detected motion-vector in described each zone of present frame detects step;
Detect the peak detection step of a plurality of peak values according to the frequency distribution of this motion-vector; With
Or/and be positioned at the information of the back frame after this present frame in time, from a plurality of peak values, judge the determination step of the motion-vector of representing present frame according to being positioned at the preceding frame before the present frame in time;
Wherein literal input determining step is according to from the output of this determination step with from the output of key operation step, selection, conversion and definite input characters.
Applications Claiming Priority (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2003101271 | 2003-04-04 | ||
| JP101271/2003 | 2003-04-04 | ||
| JP2004027880A JP3649242B2 (en) | 2003-04-04 | 2004-02-04 | Mobile phone device and character input method for mobile phone device |
| JP027880/2004 | 2004-02-04 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1536859A CN1536859A (en) | 2004-10-13 |
| CN100336375C true CN100336375C (en) | 2007-09-05 |
Family
ID=33478890
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2004100333701A Expired - Fee Related CN100336375C (en) | 2003-04-04 | 2004-04-02 | Portable terminal device and character input method |
Country Status (2)
| Country | Link |
|---|---|
| JP (1) | JP3649242B2 (en) |
| CN (1) | CN100336375C (en) |
Families Citing this family (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2531524A1 (en) | 2003-10-31 | 2005-05-12 | Iota Wireless Llc | Concurrent data entry for a portable device |
| JP4703173B2 (en) * | 2004-11-25 | 2011-06-15 | 株式会社東芝 | Mobile device |
| CN103257684B (en) | 2005-05-17 | 2017-06-09 | 高通股份有限公司 | The signal output method and device of orientation-sensitive |
| KR100675089B1 (en) | 2006-05-10 | 2007-01-29 | (주) 엘지텔레콤 | Character input method by movement of mobile terminal |
| EP1965344B1 (en) * | 2007-02-27 | 2017-06-28 | Accenture Global Services Limited | Remote object recognition |
| US8788977B2 (en) | 2008-11-20 | 2014-07-22 | Amazon Technologies, Inc. | Movement recognition as input mechanism |
| US8878773B1 (en) | 2010-05-24 | 2014-11-04 | Amazon Technologies, Inc. | Determining relative motion as input |
| US9123272B1 (en) | 2011-05-13 | 2015-09-01 | Amazon Technologies, Inc. | Realistic image lighting and shading |
| US9041734B2 (en) | 2011-07-12 | 2015-05-26 | Amazon Technologies, Inc. | Simulating three-dimensional features |
| US10088924B1 (en) | 2011-08-04 | 2018-10-02 | Amazon Technologies, Inc. | Overcoming motion effects in gesture recognition |
| US8947351B1 (en) | 2011-09-27 | 2015-02-03 | Amazon Technologies, Inc. | Point of view determinations for finger tracking |
| US9223415B1 (en) | 2012-01-17 | 2015-12-29 | Amazon Technologies, Inc. | Managing resource usage for task performance |
| US8884928B1 (en) | 2012-01-26 | 2014-11-11 | Amazon Technologies, Inc. | Correcting for parallax in electronic displays |
| US9063574B1 (en) | 2012-03-14 | 2015-06-23 | Amazon Technologies, Inc. | Motion detection systems for electronic devices |
| US9285895B1 (en) | 2012-03-28 | 2016-03-15 | Amazon Technologies, Inc. | Integrated near field sensor for display devices |
| JP6161244B2 (en) * | 2012-09-10 | 2017-07-12 | 株式会社メガチップス | Portable terminal device, program, and input method |
| US9423886B1 (en) | 2012-10-02 | 2016-08-23 | Amazon Technologies, Inc. | Sensor connectivity approaches |
| US9035874B1 (en) | 2013-03-08 | 2015-05-19 | Amazon Technologies, Inc. | Providing user input to a computing device with an eye closure |
| US9269012B2 (en) | 2013-08-22 | 2016-02-23 | Amazon Technologies, Inc. | Multi-tracker object tracking |
| US11199906B1 (en) | 2013-09-04 | 2021-12-14 | Amazon Technologies, Inc. | Global user input management |
| US10055013B2 (en) | 2013-09-17 | 2018-08-21 | Amazon Technologies, Inc. | Dynamic object tracking for user interfaces |
| US9367203B1 (en) | 2013-10-04 | 2016-06-14 | Amazon Technologies, Inc. | User interface techniques for simulating three-dimensional depth |
| JP6888351B2 (en) * | 2017-03-17 | 2021-06-16 | ヤマハ株式会社 | Input device, speech synthesizer, input method, and program |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1154015A (en) * | 1995-07-21 | 1997-07-09 | 索尼公司 | Terminal apparatus, radio communication terminal, and information input method |
| US5835641A (en) * | 1992-10-14 | 1998-11-10 | Mitsubishi Denki Kabushiki Kaisha | Image pick-up apparatus for detecting and enlarging registered objects |
| JP2000172417A (en) * | 1998-12-02 | 2000-06-23 | Nec Telecom Syst Ltd | Portable telephone device and its character input method |
| EP1136905A2 (en) * | 2000-03-17 | 2001-09-26 | Seiko Instruments Inc. | Character input device and method |
| JP2002169645A (en) * | 2000-12-01 | 2002-06-14 | Toshiba Corp | Mobile communication terminal |
| WO2002088920A1 (en) * | 2001-04-27 | 2002-11-07 | Misawa Homes Co., Ltd. | Touch-type key input apparatus |
-
2004
- 2004-02-04 JP JP2004027880A patent/JP3649242B2/en not_active Expired - Fee Related
- 2004-04-02 CN CNB2004100333701A patent/CN100336375C/en not_active Expired - Fee Related
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5835641A (en) * | 1992-10-14 | 1998-11-10 | Mitsubishi Denki Kabushiki Kaisha | Image pick-up apparatus for detecting and enlarging registered objects |
| CN1154015A (en) * | 1995-07-21 | 1997-07-09 | 索尼公司 | Terminal apparatus, radio communication terminal, and information input method |
| JP2000172417A (en) * | 1998-12-02 | 2000-06-23 | Nec Telecom Syst Ltd | Portable telephone device and its character input method |
| EP1136905A2 (en) * | 2000-03-17 | 2001-09-26 | Seiko Instruments Inc. | Character input device and method |
| JP2002169645A (en) * | 2000-12-01 | 2002-06-14 | Toshiba Corp | Mobile communication terminal |
| WO2002088920A1 (en) * | 2001-04-27 | 2002-11-07 | Misawa Homes Co., Ltd. | Touch-type key input apparatus |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1536859A (en) | 2004-10-13 |
| JP2004318826A (en) | 2004-11-11 |
| JP3649242B2 (en) | 2005-05-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN100336375C (en) | Portable terminal device and character input method | |
| CN1664761A (en) | Key-based advanced navigation techniques | |
| KR101231469B1 (en) | Method, apparatusfor supporting image processing, and computer-readable recording medium for executing the method | |
| CN101079964A (en) | Image reproducing device, image reproducing method, image reproducing program and image capturing device | |
| US8334920B2 (en) | Image playback apparatus and image display control method | |
| CN1901625A (en) | Electronic camera for capturing image as digital data | |
| CN101360189A (en) | Camera control method and camera thereof | |
| CN1617553A (en) | Apparatus and method for displaying hierarchical menu in mobile communication terminal | |
| CN1976526A (en) | Mobile telephone automation testing system and testing method | |
| JP4767200B2 (en) | Information selection device and information selection program | |
| JP2011239435A (en) | Image display unit, image display program, and image display method | |
| US9324158B2 (en) | Image processing device for performing image processing on moving image | |
| CN101076166A (en) | Device having display buttons and display method and medium for the device | |
| CN1603968A (en) | A digital camera and method for selecting focusing domain and selecting menu items | |
| CN107395926A (en) | A kind of photographic method, mobile terminal and computer-readable recording medium | |
| CN1617163A (en) | Method for selecting treating object in character identification of portable terminal and portable terminal | |
| KR101503017B1 (en) | Motion detecting method and apparatus | |
| CN108012073B (en) | Method and device for realizing panoramic shooting | |
| CN103139472B (en) | Digital photographing apparatus and its control method | |
| CN102324027A (en) | Scanning and identifying device and method | |
| JP3851601B2 (en) | Mobile terminal device | |
| JP5335474B2 (en) | Image processing apparatus and image processing method | |
| US20090058810A1 (en) | Input device and imaging device having the same | |
| KR100849532B1 (en) | Device having function of non-contact mouse and method thereof | |
| JP2008131377A (en) | Device, method and program for displaying |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20070905 Termination date: 20130402 |