JPH0410872A - Movement vector detecting device - Google Patents

Movement vector detecting device

Info

Publication number
JPH0410872A
JPH0410872A JP2113953A JP11395390A JPH0410872A JP H0410872 A JPH0410872 A JP H0410872A JP 2113953 A JP2113953 A JP 2113953A JP 11395390 A JP11395390 A JP 11395390A JP H0410872 A JPH0410872 A JP H0410872A
Authority
JP
Japan
Prior art keywords
output
circuit
movement vector
gradient component
signal
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.)
Pending
Application number
JP2113953A
Other languages
Japanese (ja)
Inventor
Toshiaki Kondo
俊明 近藤
Masayoshi Sekine
正慶 関根
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Canon Inc
Original Assignee
Canon Inc
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Canon Inc filed Critical Canon Inc
Priority to JP2113953A priority Critical patent/JPH0410872A/en
Priority to DE69121627T priority patent/DE69121627T2/en
Priority to EP91303792A priority patent/EP0454481B1/en
Publication of JPH0410872A publication Critical patent/JPH0410872A/en
Priority to US07/967,569 priority patent/US5296925A/en
Pending legal-status Critical Current

Links

Landscapes

  • Closed-Circuit Television Systems (AREA)
  • Color Television Systems (AREA)
  • Processing Of Color Television Signals (AREA)
  • Image Analysis (AREA)
  • Studio Circuits (AREA)

Abstract

PURPOSE:To eliminate accuracy degradation and to attain real time processing by operating a movement vector according to an estimation expression to which the effect of an inclined directional space gradient component is approximated and finding a signal to represent the inclined directional space gradient component quantity of an input picture signal. CONSTITUTION:Summing circuits 32, 34 and 36 to accumulate-add the output data of a multiplier 26, an absolute value circuit 30 and a multiplier 28 respectively within a designated block and a divider 38 to divide the output of the summing circuit 32 with the output of the summing circuit 34 are provided. Then, a first operation to calculate a movement vector according to an estimation expression to which the effect of an inclined directional space gradient component is approximated is executed, a second operation to find a signal to represent the inclined directional space gradient component quantity of an input picture signal is executed and moreover, a third operation to average the output of the result of the first operation under a weight according to the output of the result of the second operation is executed. Thus, the movement vector can be detected at high speed with high accuracy.

Description

【発明の詳細な説明】 [産業上の利用分野コ 本発明は動きベクトル検出装置に関し、より具体的には
、TV左カメラ電子カメラ、ビデオ・カメラ及び工業用
画像計測機器などの撮像光学装置、特に防振や追尾など
の機能を有する撮像光学装置における動きベクトル検出
装置に関する。
DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a motion vector detection device, and more specifically, to an imaging optical device such as a TV left camera electronic camera, a video camera, and an industrial image measuring device; In particular, the present invention relates to a motion vector detection device in an imaging optical device having functions such as image stabilization and tracking.

[従来の技術] 画像信号処理による動きベクトル検出法としては、昭和
60年特許出願公告第46878号や米国特許第389
0462号に記載される時空間勾配法がある。
[Prior Art] As a motion vector detection method using image signal processing, there are methods such as those disclosed in Patent Application Publication No. 46878 of 1985 and U.S. Patent No. 389.
There is a spatio-temporal gradient method described in No. 0462.

X方向の画像変位量をα、X方向の画像変位量をβとす
るとき、前者では、 α=(Σ6g、゛″(Σagy ’d)−(Σeg、’
g、’)(Σngy’d)1/(Σtrgx ’ 2Σ
Bgy”べΣag−’gy’)2)β=(Σig−”(
Σsgy’d)−(Σl1g、’gy’)(ΣBg%d
)1/(Σegg ”Σigy”−(ΣBg、’gy’
)’)    (1)で与えられ、後者では、 α=ΣHd−sign(g、’)/Σm l g。
When the amount of image displacement in the X direction is α and the amount of image displacement in the
g,')(Σngy'd)1/(Σtrgx' 2Σ
Bgy"beΣag-'gy')2) β=(Σig-"(
Σsgy'd) - (Σl1g,'gy')(ΣBg%d
)1/(Σegg "Σigy" - (ΣBg, 'gy'
)') (1), in the latter, α=ΣHd−sign(g,′)/Σm l g.

β=Σ、 d−sign(gy’)/Σs1gy’l 
  (2)で与えられる。但し、dは時間的に連続する
画像間の同じ位置における濃度(レベル)差、即ち時間
勾配を示し、g−’1gy’はそれぞれ画像をgで表わ
したときのX方向、X方向の空間勾配を示す。
β=Σ, d-sign(gy')/Σs1gy'l
It is given by (2). However, d indicates the density (level) difference at the same position between temporally consecutive images, that is, the temporal gradient, and g-'1gy' indicates the spatial gradient in the X direction and the X direction, respectively, when the image is represented by g. shows.

また、Σ8は、ブロック内の総和演算を意味し、Sig
n()はg、’、 gy’の符号を出力する関数である
Also, Σ8 means the summation operation within the block, and Sig
n() is a function that outputs the signs of g, ', gy'.

高速の演算が要求される実時間画像処理では、式(2)
が注目されている。
In real-time image processing that requires high-speed calculation, Equation (2)
is attracting attention.

[発明が解決しようとする課題] しかし、式(2)は、式(1)のΣBgx’gy’の項
をゼロと近似して式(1)を簡略化してものであるので
、ΣBg−’gy’が大きな値になるときには、精度が
著しく劣化するという欠点がある。
[Problem to be solved by the invention] However, since equation (2) is a simplified version of equation (1) by approximating the term ΣBgx'gy' in equation (1) to zero, ΣBg-' When gy' becomes a large value, there is a drawback that the accuracy deteriorates significantly.

そこで本発明は、このような精度劣化がなく、しかも実
時間処理に適した動きベクトル検出装置を提示すること
を目的とする。
Therefore, it is an object of the present invention to provide a motion vector detection device that does not suffer from such deterioration in accuracy and is suitable for real-time processing.

[課題を解決するための手段] 本発明に係る動きベクトル検出装置は、画面間の濃度差
及び画面内の空間勾配から画像の動き量を求める動きベ
クトル検出装置であって、斜め方向空間勾配成分の影響
を近似した推定式に従い動きベクトルを演算する第1の
演算手段と、入力画像信号の斜め方向空間勾配成分量を
表わす信号を求める第2の演算手段と、第1の演算手段
の出力を第2の演算手段の出力に従う重みのもとで平均
化する第3の演算手段とからなることを特徴とする。
[Means for Solving the Problems] A motion vector detection device according to the present invention is a motion vector detection device that calculates the amount of movement of an image from the density difference between screens and the spatial gradient within the screen, and which calculates the amount of movement of an image from the density difference between screens and the spatial gradient within the screen. a first calculation means for calculating a motion vector according to an estimation formula that approximates the influence of the input image signal; a second calculation means for calculating a signal representing the diagonal spatial gradient component amount of the input image signal; and a third calculation means that performs averaging under a weight according to the output of the second calculation means.

[作用] 上記第1の演算手段により、基本的には近似推定式によ
り動きベクトルを求めているので、高速である。また、
上記第2及び第3の演算手段により斜め方向空間勾配成
分量を考慮しているので、高い信頼性、即ち精度を得る
ことができる。
[Operation] The first calculation means basically calculates the motion vector using an approximate estimation formula, so it is fast. Also,
Since the second and third calculation means take into consideration the spatial gradient component amount in the diagonal direction, high reliability, that is, accuracy can be obtained.

[実施例] 以下、図面を参照して本発明の詳細な説明する。[Example] Hereinafter, the present invention will be described in detail with reference to the drawings.

第1図は本発明の一実施例の構成ブロック図を示す。1
0は画像信号の入力端子、12は1フイールド(又はフ
レーム)期間、入力信号を記憶(即ち、時間遅延)する
レジスタ、14は入力端子10の画像信号からレジスタ
12の出力を減算する減算器、16は画像濃度分布の空
間勾配を演算するのに必要な数画素分の走査時間だけ、
その入力信号を記憶(即ち、時間遅延)するレジスタ、
18は入力端子10の画像信号からレジスタ16の出力
を減算する減算器、20はレジスタ16及び減算器18
により求められる空間勾配と直交する方向の空間勾配を
求めるために、所定走査期間だけ入力端子10の画像信
号を記憶(即ち、時間遅延)するレジスタ、22は入力
端子10の画像信号からレジスタ20の出力を減算する
減算器である。
FIG. 1 shows a block diagram of an embodiment of the present invention. 1
0 is an input terminal for the image signal; 12 is a register that stores the input signal for one field (or frame) period (i.e., time delay); 14 is a subtracter that subtracts the output of the register 12 from the image signal of the input terminal 10; 16 is the scanning time for several pixels necessary to calculate the spatial gradient of the image density distribution;
a register that stores (i.e., time delays) the input signal;
18 is a subtracter that subtracts the output of the register 16 from the image signal of the input terminal 10; 20 is the register 16 and the subtracter 18;
In order to obtain a spatial gradient in a direction perpendicular to the spatial gradient obtained by It is a subtracter that subtracts the output.

24は減算器18の出力の符号(正、負又はセロ)を示
す信号を出力する符号出力回路、26は、減算器14の
出力に符号出力回路24から出力される符号を乗算する
乗算器、28は減算器18の出力に減算器22の出力を
乗算する乗算器、30は減算器24の出力の絶対値を出
力する絶対値回路、32,34.36は指定ブロック内
で、それぞれ乗算器32、絶対値回路30及び乗算器2
8の出力データを累積加算する総和回路、38は総和回
路32の出力を総和回路34の出力で除算する除算器で
ある。除算器38の出力は各ブロックの動きベクトルで
ある。40は除算器38から出力される動きベクトルを
記憶するメモリ、42は総和回路36の出力の大きさ順
に、メモリ40か記憶する動きベクトルを整列(ソート
)する整列回路、44は整列回路42により整列された
動きベクトルを適当な重みを付けて平均化する重み付き
平均化回路、46は動きベクトルの最終的な出力端子で
ある。
24 is a sign output circuit that outputs a signal indicating the sign (positive, negative, or zero) of the output of the subtracter 18; 26 is a multiplier that multiplies the output of the subtracter 14 by the sign output from the sign output circuit 24; 28 is a multiplier that multiplies the output of subtracter 18 by the output of subtracter 22; 30 is an absolute value circuit that outputs the absolute value of the output of subtracter 24; 32, 34, and 36 are multipliers in the designated block, respectively. 32, absolute value circuit 30 and multiplier 2
38 is a divider that divides the output of the summation circuit 32 by the output of the summation circuit 34. The output of divider 38 is the motion vector for each block. 40 is a memory that stores the motion vectors output from the divider 38; 42 is a sorting circuit that sorts (sorts) the motion vectors stored in the memory 40 in order of the magnitude of the output of the summation circuit 36; 44 is a sorting circuit that sorts the motion vectors stored in the memory 40; A weighted averaging circuit 46 which averages the aligned motion vectors with appropriate weights is the final output terminal of the motion vectors.

第1図の動作を説明する。入力端子10に入力した画像
信号gは3経路に分けられる。即ち、レジスタ12及び
減算器14により、時間的に連続する2つのフィールド
(又はフレーム)画面間での濃度差、即ち時間勾配dが
算出され、レジスタ16及び減算器18により、前フィ
ールド(又はフレーム)画面内のX方向の空間勾配g*
′か算出され、レジスタ20及び減算器22により、前
フ内 イールド(又はフレーム)画面謄4のX方向の空間勾配
g、“が算出される。
The operation shown in FIG. 1 will be explained. The image signal g input to the input terminal 10 is divided into three paths. That is, the register 12 and the subtracter 14 calculate the density difference between two temporally consecutive field (or frame) screens, that is, the temporal gradient d, and the register 16 and the subtracter 18 calculate the density difference between two temporally consecutive fields (or frames), and the register 16 and the subtracter ) Spatial gradient g* in the X direction within the screen
The register 20 and the subtracter 22 calculate the spatial gradient g, ``of the previous field yield (or frame) screen 4 in the X direction.

X方向の空間勾配gx′は、符号出力回路24、絶対値
回路30及び乗算器28に印加され、符号出力回路24
は空間勾配訃°か正のときには、十1、ゼロのときには
0、負のときには−1を出力し、絶対値回路30は空間
勾配g8′の絶対値を出力する。乗算器26は時間勾配
d(減算器14の出力)に符号出力回路24の出力を乗
算する。そして、総和回路32.34はそれぞれ、所定
数の画素からなるブロック内で、乗算器26及び絶対値
回路30の出力の総和をとる。即ち、総和回路32はΣ
、 d−sign(g%)を計算し、総和回路34はΣ
m1g−’lを計算する。除算器38は総和回路32の
出力を総和回路34の出力で除算する。除算器38の出
力は、前述の式(2)のαに相当する。
The spatial gradient gx' in the X direction is applied to the sign output circuit 24, the absolute value circuit 30, and the multiplier 28, and the sign output circuit 24
outputs 11 when the spatial gradient g8' is positive, 0 when it is zero, and -1 when it is negative, and the absolute value circuit 30 outputs the absolute value of the spatial gradient g8'. The multiplier 26 multiplies the time gradient d (output of the subtracter 14) by the output of the sign output circuit 24. The summation circuits 32 and 34 each sum the outputs of the multiplier 26 and the absolute value circuit 30 within a block consisting of a predetermined number of pixels. That is, the summation circuit 32 is Σ
, d-sign (g%), and the summation circuit 34 calculates Σ
Calculate m1g-'l. A divider 38 divides the output of the summation circuit 32 by the output of the summation circuit 34. The output of the divider 38 corresponds to α in equation (2) above.

ところで乗算器28は、X方向の空間勾配g。By the way, the multiplier 28 calculates the spatial gradient g in the X direction.

にX方向の空間勾配gy’を乗算し、総和回路36は総
和回路32.34と同様のブロックで、乗算器28の出
力の総和を計算する。総和回路36の出力Σ5g、′・
gy’は、式(1)を式(2)の形に簡略化するために
ゼロと近似した項である。式(2)により求めた動きベ
クトル(除算器38の出力)の信頼性又は確実性を、総
和回路36の出力により判別できる。即ち、総和回路3
6の出力か充分にゼロに近ければ、式(2)による推定
結果は信頼性か高く、逆に大きい値であれば信頼性か低
いということになる。
is multiplied by the spatial gradient gy' in the X direction, and the summation circuit 36 is a block similar to the summation circuits 32 and 34, and calculates the sum of the outputs of the multiplier 28. The output of the summation circuit 36 Σ5g,'・
gy' is a term approximated to zero in order to simplify equation (1) into equation (2). The reliability or certainty of the motion vector (output of the divider 38) obtained by equation (2) can be determined based on the output of the summation circuit 36. That is, summation circuit 3
If the output of 6 is sufficiently close to zero, the estimation result based on equation (2) is highly reliable, and conversely, if the output is large, the reliability is low.

そこで、除算器38により得られた動きベクトルを一旦
、メモリ40に格納し、整列回路42によりΣmg、x
 ’・g、°の小さい順に整列し、重み付は平均化回路
44では、Σ8g!°・g、゛が小さいはと大きな重み
を付けて平均化を行なう。これにより、出力端子46か
らは、より信頼性の高い画像移動量の信号を得ることが
できる。
Therefore, the motion vector obtained by the divider 38 is temporarily stored in the memory 40, and the alignment circuit 42 calculates Σmg, x
'·g, ° are arranged in ascending order, and weighted by the averaging circuit 44, Σ8g! Averaging is performed by giving a larger weight to the smaller °·g, ゛. Thereby, a more reliable image movement amount signal can be obtained from the output terminal 46.

次に、第2図を参照して本実施例の効果を説明する。第
2図は、固体撮像素子のカメラにより得られた水平×垂
直×階Q=256X256X 8ビツトの画像信号に本
実施例を適用した場合のシミュレーション結果である。
Next, the effects of this embodiment will be explained with reference to FIG. FIG. 2 shows the simulation results when this embodiment is applied to an image signal of horizontal x vertical x floor Q = 256 x 256 x 8 bits obtained by a camera of a solid-state image sensor.

画像が水平方向に1〜10画素移動したときに、垂直方
向の推定結果を示す。垂直方向には移動していないので
、垂直方向の推定結果は、いわば、誤検出量である。第
3図は、画像を水平方向にのみ1〜10画素移動させた
場合の、画像の水平移動量に対する垂直方向の推定結果
を第3図に図示した。第3図で、○か本実施例の誤検出
量、×か従来の勾配法による誤検出量である。
The estimation results in the vertical direction are shown when the image moves horizontally by 1 to 10 pixels. Since there is no movement in the vertical direction, the estimation result in the vertical direction is, so to speak, an amount of false detection. FIG. 3 shows the estimation results in the vertical direction relative to the amount of horizontal movement of the image when the image is moved only in the horizontal direction by 1 to 10 pixels. In FIG. 3, ○ indicates the amount of false detection in this embodiment, and × indicates the amount of false detection by the conventional gradient method.

第3図から分かるように、Σ1g、′・g、°による重
み付は平均化により、誤検出量が格段に減少している。
As can be seen from FIG. 3, the weighting by Σ1g,'·g,° significantly reduces the amount of false detections by averaging.

なお、このシミュレーションでは、1画面のブロック数
を16とし、各ブロックで得られた動きベクトルを評価
関数Σ3g8′・gy’の順に整列した後、この順で1
6から順に1ずつ減少する重みを乗せて平均化を行なっ
た。
In this simulation, the number of blocks in one screen is 16, and the motion vectors obtained in each block are arranged in the order of the evaluation functions Σ3g8' and gy', and then 1
Averaging was performed by applying weights that decreased by 1 starting from 6.

重み付き平均化回路44における重み付けの方法として
は、除算器38の出力から得られる動きベクトルに評価
関数Σmgx ’・gy’の逆数を乗算する方法であっ
てもよい。そうすれば、評価関数Σ9g・g、°の値そ
のものが反映されるので、動きベクトルの信頼性が更に
向上する。勿論、この場合には、整列回路42は無くて
もよい。
The weighting method in the weighted averaging circuit 44 may be a method in which the motion vector obtained from the output of the divider 38 is multiplied by the reciprocal of the evaluation function Σmgx'·gy'. By doing so, the value of the evaluation function Σ9g·g,° itself is reflected, so the reliability of the motion vector is further improved. Of course, in this case, the alignment circuit 42 may be omitted.

第3図は第1図の変更実施例の構成ブロック図を示す。FIG. 3 shows a block diagram of a modified embodiment of FIG. 1.

但し、第1図と同じ構成要素には同し符号を付しである
。変更部分を説明する。48,50は総和回路32.3
4と同様の総和回路であり、それぞれ減算器18.22
の出力の、ブロック毎の総和を計算する。52.54は
総和回路4850の出力の絶対値をとり、加算器56は
絶対値回路52.54の出力を加算する。整列回路58
は、加算器56の出ノJの大きい順に、メモリ40の動
きベクトルを整列し、重み付は平均化回路60は整列回
路58による整列順に重みを付けて平均化し、出力端子
62に出力する。
However, the same components as in FIG. 1 are given the same reference numerals. Explain the changed parts. 48, 50 are summation circuits 32.3
It is a summation circuit similar to 4, and has subtracters 18 and 22 respectively.
Calculate the sum of the outputs for each block. 52.54 takes the absolute value of the output of the summation circuit 4850, and the adder 56 adds the outputs of the absolute value circuit 52.54. Alignment circuit 58
The motion vectors in the memory 40 are arranged in descending order of the output J of the adder 56, and the weighted averaging circuit 60 weights and averages them in the order of sorting by the arrangement circuit 58, and outputs the averaged vectors to an output terminal 62.

加算器56の出力は1Σ5gff’l+lΣ@gy’l
であり、斜め方向の勾配成分量を表わしている。重み付
は平均化回路60は、斜め成分の大きなブロックの動き
ベクトルを低く評価し、斜め成分の小さいなブロックの
動きベクトルを高く評価する。
The output of the adder 56 is 1Σ5gff'l+lΣ@gy'l
, which represents the amount of gradient component in the diagonal direction. The weighting averaging circuit 60 evaluates motion vectors of blocks with large diagonal components low, and highly evaluates motion vectors of blocks with small diagonal components.

但し、第3図の実施例では、]Σag、’lと]21g
と間に大きな値の差かあると、例えば大小比率が20倍
を越える場合、それは水平方向あるいは垂直方向への1
方向の空間勾配があることを示しており、実際には斜め
方向成分か小さいことになる。従って、重み付は平均化
回路6oではこの点を考慮した重み付けを行なう必要か
ある。
However, in the embodiment of FIG. 3, ]Σag, 'l and ]21g
If there is a large value difference between
This indicates that there is a spatial gradient in the direction, and in reality the diagonal component is small. Therefore, it is necessary to perform weighting in consideration of this point in the averaging circuit 6o.

[発明の効果コ 以上の説明から容易に理解できるように、本発明によれ
ば、斜め成分による検出精度の劣化を抑制でき、動きベ
クトルを高速且つ高精度に検出できる。
[Effects of the Invention] As can be easily understood from the above description, according to the present invention, deterioration in detection accuracy due to oblique components can be suppressed, and motion vectors can be detected at high speed and with high accuracy.

【図面の簡単な説明】[Brief explanation of the drawing]

第1図は本発明の一実施例の構成ブロック図、第2図は
第1図及び従来例による水平方向移動量に対する垂直方
向の推定結果を示す比較図、第3図は本発明の変更実施
例の構成ブロック図である。 10:入力端子 12,16;  20:レジスタ14
.18,22:減算器 24:符号出力回路26.28
:乗算器 30,52,54:絶対値回路 32,34
,36,48,50:総和回路 38;除算器 40:
メモリ 42.・58:整列回路 44,60:重み付
は平均化回路 46.62:出力端子 56:加算器 箱3 s+mhae mz−n四
FIG. 1 is a block diagram of the configuration of an embodiment of the present invention, FIG. 2 is a comparison diagram showing estimation results in the vertical direction relative to the amount of horizontal movement according to FIG. 1 and the conventional example, and FIG. 3 is a modified implementation of the present invention. FIG. 2 is a configuration block diagram of an example. 10: Input terminal 12, 16; 20: Register 14
.. 18, 22: Subtractor 24: Sign output circuit 26.28
: Multiplier 30, 52, 54: Absolute value circuit 32, 34
, 36, 48, 50: summation circuit 38; divider 40:
Memory 42.・58: Sorting circuit 44, 60: Weighting is averaging circuit 46.62: Output terminal 56: Adder box 3 s+mhae mz-n 4

Claims (1)

【特許請求の範囲】[Claims] 画面間の濃度差及び画面内の空間勾配から画像の動き量
を求める動きベクトル検出装置であって、斜め方向空間
勾配成分の影響を近似した推定式に従い動きベクトルを
演算する第1の演算手段と、入力画像信号の斜め方向空
間勾配成分量を表わす信号を求める第2の演算手段と、
第1の演算手段の出力を第2の演算手段の出力に従う重
みのもとで平均化する第3の演算手段とからなることを
特徴とする動きベクトル検出装置。
A motion vector detection device that calculates the amount of movement of an image from a density difference between screens and a spatial gradient within the screen, the device comprising: a first calculation means that calculates a motion vector according to an estimation formula that approximates the influence of a diagonal spatial gradient component; , second calculation means for calculating a signal representing the amount of the diagonal spatial gradient component of the input image signal;
A motion vector detection device comprising: a third calculation means that averages the output of the first calculation means under a weight according to the output of the second calculation means.
JP2113953A 1990-04-27 1990-04-27 Movement vector detecting device Pending JPH0410872A (en)

Priority Applications (4)

Application Number Priority Date Filing Date Title
JP2113953A JPH0410872A (en) 1990-04-27 1990-04-27 Movement vector detecting device
DE69121627T DE69121627T2 (en) 1990-04-27 1991-04-26 Device for the detection of motion vector
EP91303792A EP0454481B1 (en) 1990-04-27 1991-04-26 Movement vector detection device
US07/967,569 US5296925A (en) 1990-04-27 1992-10-27 Movement vector detection device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP2113953A JPH0410872A (en) 1990-04-27 1990-04-27 Movement vector detecting device

Publications (1)

Publication Number Publication Date
JPH0410872A true JPH0410872A (en) 1992-01-16

Family

ID=14625347

Family Applications (1)

Application Number Title Priority Date Filing Date
JP2113953A Pending JPH0410872A (en) 1990-04-27 1990-04-27 Movement vector detecting device

Country Status (1)

Country Link
JP (1) JPH0410872A (en)

Similar Documents

Publication Publication Date Title
KR930002613B1 (en) Image motion vector detector
US7346109B2 (en) Motion vector computation for video sequences
US4695882A (en) Movement estimation system for video signals using a recursive gradient method
US7536031B2 (en) Temporal interpolation of a pixel on basis of occlusion detection
US20100271554A1 (en) Method And Apparatus For Motion Estimation In Video Image Data
JP4053422B2 (en) Method for reducing the influence of halo in interpolation by motion compensation
US7499494B2 (en) Vector selection decision for pixel interpolation
JPH06197264A (en) Digital image stabilization method and system
KR0166724B1 (en) Apparatus and method for estimating a half-pel motion vector
CA2289948C (en) Motion vector processing circuit
KR100217485B1 (en) Method for movement compensation in a moving-image encoder or decoder
GB2165417A (en) Measurement and correction of film unsteadiness in a video signal
JPH043595A (en) Moving information detection system for moving picture and prediction coding system for inter-frame moving compensation for moving picture
US5943450A (en) Apparatus and method for compensating for camera vibration during video photography
JP2969781B2 (en) Motion vector detection device
JPH0410872A (en) Movement vector detecting device
US5173770A (en) Movement vector detection device
JP3627872B2 (en) Motion vector detection method and apparatus
JP3252418B2 (en) Image shake determination device
JP3003044B2 (en) Apparatus and method for detecting motion vector
JP2600520B2 (en) Image motion compensation device
JPH0410874A (en) Movement vector detecting device
JP2975146B2 (en) Motion vector detection device
JPH05176218A (en) Picture motion correcting device
JP2910284B2 (en) Moving vector detecting method and apparatus