JPH0815401A - Image tracking apparatus - Google Patents
Image tracking apparatusInfo
- Publication number
- JPH0815401A JPH0815401A JP6147880A JP14788094A JPH0815401A JP H0815401 A JPH0815401 A JP H0815401A JP 6147880 A JP6147880 A JP 6147880A JP 14788094 A JP14788094 A JP 14788094A JP H0815401 A JPH0815401 A JP H0815401A
- Authority
- JP
- Japan
- Prior art keywords
- area
- output
- tracking
- measuring
- region
- 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
Links
Landscapes
- Aiming, Guidance, Guns With A Light Source, Armor, Camouflage, And Targets (AREA)
- Image Processing (AREA)
- Closed-Circuit Television Systems (AREA)
- Image Analysis (AREA)
Abstract
Description
【０００１】[0001]
【産業上の利用分野】この発明は、画像追尾装置に関す
るものである。BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image tracking device.
【０００２】[0002]
【従来の技術】図１３は従来の画像追尾装置の構成図で
ある。図１３において、１は撮像機、２は二値化手段、
３は領域毎重心計測手段、４は重心位置誤差計測手段、
５は目標選択手段、６は追尾位置出力である。2. Description of the Related Art FIG. 13 is a block diagram of a conventional image tracking device. In FIG. 13, 1 is an image pickup device, 2 is a binarization unit,
3 is a centroid measuring means for each area, 4 is a centroid position error measuring means,
Reference numeral 5 is a target selecting means, and 6 is a tracking position output.
【０００３】従来の画像追尾装置は、撮像機１により濃
淡画像を得る。二値化手段２は前記濃淡画像を“数１”
により二値化する。“数１”において、ｇ（ｘ、ｙ）は
濃淡画像の座標（ｘ、ｙ）の画素の輝度、Ｔはしきい
値、ｆ（ｘ、ｙ）は二値画像の座標（ｘ、ｙ）の画素値
で１ならば有意画素である。領域毎重心計測手段３は前
記二値化手段２により得られた二値画像において有意画
素が連結している領域毎にその領域の重心位置を計測す
る。重心位置誤差計測手段４は、今回検出した領域の重
心位置と前回目標領域と定めた領域の重心位置の誤差を
計測する。目標選択手段５は、前記重心位置誤差計測手
段４の出力する誤差が許容範囲内で最小となる領域を目
標領域として選出する。許容範囲は重心位置の変動が大
き過ぎるものを除外するためあらかじめ決定しておく。
このようにして選出された領域の重心位置を今回の追尾
位置出力６として出力する。In the conventional image tracking device, a grayscale image is obtained by the image pickup device 1. The binarizing means 2 converts the grayscale image into "Equation 1".
Binarize by. In “Equation 1”, g (x, y) is the luminance of the pixel at the coordinates (x, y) of the grayscale image, T is a threshold value, and f (x, y) is the coordinates (x, y) of the binary image. If the pixel value of 1 is 1, the pixel is significant. The region-by-region centroid measuring unit 3 measures the centroid position of each region in the binary image obtained by the binarizing unit 2, in which significant pixels are connected. The center-of-gravity position error measuring means 4 measures the error between the center-of-gravity position of the region detected this time and the center-of-gravity position of the region defined as the previous target region. The target selecting means 5 selects, as a target area, an area in which the error output from the barycentric position error measuring means 4 is the smallest within the allowable range. The permissible range is determined in advance in order to exclude those in which the variation of the position of the center of gravity is too large.
The center-of-gravity position of the area thus selected is output as the tracking position output 6 of this time.
【０００４】[0004]
【数１】 [Equation 1]
【０００５】[0005]
【発明が解決しようとする課題】艦船では、甲板は太陽
光線があたり明るく温められていて、構造物の陰などは
暗く冷えているため、撮像機で撮像された濃淡画像上で
は甲板は輝度が高く、構造物の陰などは輝度が低くな
る。艦船あるいは撮像機側の移動により、それまで見え
なかった構造物の陰が見えるようになると、輝度が低く
二値化で有意画素とならない部分で領域が分割すること
がある。図１４は目標領域が分割する前後の模式的な二
値画像である。従来の画像追尾装置は上記のように構成
されているので、目標を抽出し追尾する場合に、図１４
（ａ）に示すように１つの領域を追尾中、図１４（ｂ）
に示すように目標領域が２つ以上に分割すると、分割し
た領域の重心位置と分割前の追尾位置との誤差が大きく
なり目標選択の対象外となり、追尾継続できないという
難点があった。SUMMARY OF THE INVENTION In a ship, the deck is warmed brightly by the sun's rays, and the shade of the structure is dark and cold. It is high, and the shadow of the structure has low brightness. When the shadow of a structure that was not visible until then becomes visible due to the movement of the ship or the imaging device side, the region may be divided at a portion where the brightness is low and the pixel does not become a significant pixel by binarization. FIG. 14 is a schematic binary image before and after the target area is divided. Since the conventional image tracking device is configured as described above, when the target is extracted and tracked, as shown in FIG.
Tracking one area as shown in FIG.
When the target area is divided into two or more as shown in (3), the error between the center of gravity of the divided area and the tracking position before the division becomes large, and the target is not selected.
【０００６】この発明は、このような課題を解消するた
めになされたもので、追尾中の１つの領域が追尾途中で
２つ以上に分割しても目標として追尾継続できる画像追
尾装置を得ることを目的とする。The present invention has been made in order to solve such a problem, and provides an image tracking device capable of continuing tracking as a target even if one region under tracking is divided into two or more in the middle of tracking. With the goal.
【０００７】[0007]
【課題を解決するための手段】この発明に係る画像追尾
装置は、二値画像の領域毎の面積を計測する領域毎面積
計測手段、二値画像の各領域の重心を計測する領域毎重
心計測手段、二値画像の各領域の主軸を計測する領域毎
主軸計測手段、二値画像を１画面分遅らせて出力するフ
レームメモリと、前記のフレームメモリの出力する前フ
レームの二値画像と前フレームでの目標候補を指定する
領域番号とから二値画像中の領域番号に対応する全ての
領域について主軸を計測する追尾領域主軸計測手段と、
領域毎重心計測手段の出力と追尾領域主軸計測手段の出
力を受けて領域毎の重心と前回の追尾領域の主軸との距
離を計測する重心−主軸間距離計測手段、領域毎主軸計
測手段の出力と追尾領域主軸計測手段の出力を受けて領
域毎の主軸と前回の追尾領域の主軸のなす角を計測する
主軸傾き誤差計測手段、目標を同定する手段として、重
心−主軸間距離計測手段の出力と主軸傾き誤差計測手段
の出力を受けて重心−主軸間距離と主軸方向誤差がそれ
ぞれ許容値以内である領域すべてを今回の目標候補と
し、その領域番号を出力する目標候補選択手段、目標候
補選択手段の出力する複数の目標候補領域の面積と重心
から各目標候補領域の合成重心位置を追尾位置として出
力する追尾位置計測手段を設けたものである。An image tracking apparatus according to the present invention is a region-by-region area measuring means for measuring an area for each region of a binary image, and a region-by-region gravity center measurement for measuring a gravity center of each region of a binary image. Area measuring means for measuring the main axis of each area of the binary image, frame memory for outputting the binary image delayed by one screen, and binary image of the previous frame and previous frame output by the frame memory Tracking area spindle measuring means for measuring the spindle for all areas corresponding to the area number in the binary image from the area number designating the target candidate in
The distance between the center of gravity of each region and the distance between the center of gravity of each region and the spindle of the previous tracking region is measured by receiving the output of the center of gravity measuring device for each region and the output of the tracking region spindle measuring device, and the output of the region measuring spindle measuring device. And an output of the center-of-spindle distance measuring means as a means for identifying a target, a spindle inclination error measuring means for measuring the angle between the spindle for each area and the spindle of the previous tracking area in response to the output of the tracking area spindle measuring means. In response to the output of the spindle inclination error measuring means, all the areas in which the distance between the center of gravity and the spindle and the error in the spindle direction are within the allowable values are set as the target candidates for this time, and the candidate number selecting means and the target candidate selecting means that output the area number. Tracking position measuring means is provided for outputting, as a tracking position, the composite barycentric position of each target candidate area from the area and the center of gravity of the plurality of target candidate areas output by the means.
【０００８】またこの発明は、二値画像の各領域の面積
を計測する領域毎面積計測手段、二値画像の各領域の重
心を計測する領域毎重心計測手段、二値画像を１画面分
遅らせて出力する第１のフレームメモリと、前記の第１
のフレームメモリの出力と目標候補を指定する領域番号
を受けて第１のフレームメモリの出力する二値画像の内
目標候補領域番号で指定された領域の画素のみ有意と
し、他の画素は有意でないと判断する目標領域画素判定
手段と、前記の目標領域画素判定手段の出力する二値画
像を記憶する第２のフレームメモリと、領域毎重心位置
計測手段の出力する各領域の重心位置が第２のフレーム
メモリの二値画像において有意画素か判定し有意画素で
あれば目標候補領域とし、その領域番号を出力する重心
位置判定による目標候補選択手段、重心位置判定による
目標候補選択手段の出力する複数の目標候補領域の面積
と重心から各目標候補領域の合成重心位置を追尾位置と
して出力する追尾位置計測手段を設けたものである。Further, according to the present invention, the area-by-area area measuring means for measuring the area of each area of the binary image, the area-by-area center of gravity measuring means for measuring the center of gravity of each area of the binary image, and the binary image being delayed by one screen. To output the first frame memory and the first frame memory
When the output of the frame memory and the area number designating the target candidate are received, only the pixels of the area designated by the target candidate area number of the binary image output from the first frame memory are significant, and the other pixels are not significant. Target area pixel determining means, a second frame memory that stores the binary image output by the target area pixel determining means, and the center-of-gravity position of each area output by the area-by-area center-of-gravity position measuring means is second. In the binary image of the frame memory, if a significant pixel is determined, and if it is a significant pixel, it is set as a target candidate area, and the area number is output. Tracking position measuring means for outputting the combined barycentric position of each target candidate area as the tracking position from the area and the center of gravity of the target candidate area is provided.
【０００９】この発明は、二値画像の各領域の重心を計
測する領域毎重心計測手段、二値画像の各領域の主軸を
計測する領域毎主軸計測手段、二値画像を１画面分遅ら
せて出力する第１のフレームメモリと、前記の第１のフ
レームメモリの出力する前フレームの二値画像と前フレ
ームでの目標候補を指定する領域番号とから二値画像中
の領域番号に対応する全ての領域について主軸を計測す
る追尾領域主軸計測手段と、領域毎重心計測手段の出力
と追尾領域主軸計測手段の出力を受けて領域毎の重心と
前回の追尾領域の主軸との距離を計測する重心−主軸間
距離計測手段、領域毎主軸計測手段の出力と追尾領域主
軸計測手段の出力を受けて領域毎の主軸と前回の追尾領
域の主軸のなす角を計測する主軸傾き誤差計測手段、目
標を同定する手段として、重心−主軸間距離計測手段の
出力と主軸傾き誤差計測手段の出力を受けて重心−主軸
間距離と主軸方向誤差がそれぞれ許容値以内である領域
すべてを今回の目標候補とし、その領域番号を出力する
目標候補選択手段、前記の第１のフレームメモリの出力
と目標候補を指定する領域番号を受けて第１のフレーム
メモリの出力する二値画像の内目標候補領域番号で指定
された領域の画素のみ有意とし、他の画素は有意でない
と判断する目標領域画素判定手段と、前記の目標領域画
素判定手段の出力する二値画像を記憶する第２のフレー
ムメモリと、前記の第２のフレームメモリの出力を受け
て前回の目標候補領域の二値画像を記憶する第３のフレ
ームメモリと、目標候補選択手段で選択された複数の目
標候補領域から追尾位置を計測する方法として第２のフ
レームメモリと第３のフレームメモリの画素毎の論理和
演算結果画像の重心位置を追尾位置として出力する領域
論理和による追尾位置計測手段を設けたものである。According to the present invention, the centroid measuring means for each area for measuring the centroid of each area of the binary image, the spindle measuring means for each area for measuring the spindle of each area of the binary image, and the binary image delayed by one screen. From the first frame memory to output, the binary image of the previous frame output from the first frame memory, and the area number designating the target candidate in the previous frame, all corresponding to the area number in the binary image Tracking area main axis measuring means for measuring the main axis of the area, and the center of gravity for measuring the distance between the center of gravity of each area and the previous tracking area main axis by receiving the outputs of the area center of gravity measuring means and the tracking area main axis measuring means. A spindle tilt error measuring means for measuring the angle between the spindle for each area and the spindle of the previous tracking area, receiving the output of the spindle distance measuring means for each spindle and the output of the tracking area spindle measuring means for the target; Means to identify Then, receiving the output of the center of gravity-spindle distance measuring means and the output of the spindle tilt error measuring means, all the areas in which the center of gravity-spindle distance and the spindle direction error are within the respective allowable values are set as the target candidates for this time, and that area is set. A target candidate selecting means for outputting a number, the output of the first frame memory and the area number for specifying the target candidate are received, and the target candidate area number of the binary image output by the first frame memory is designated. Target area pixel determining means for determining that only the pixels of the area are significant and other pixels are not significant, a second frame memory for storing the binary image output by the target area pixel determining means, and the second The tracking position is measured from the third frame memory that receives the output of the frame memory of the target candidate area and stores the binary image of the previous target candidate area and the plurality of target candidate areas selected by the target candidate selecting means. Is provided with a second frame memory and the third tracking position measuring means by the region ORing the center of gravity of the logical OR operation result image of each pixel of the frame memory is output as the tracking position modulo.
【００１０】この発明は、二値画像の各領域の面積を計
測する領域毎面積計測手段、二値画像の領域毎の重心を
計測する領域毎重心計測手段、二値画像を１画面分遅ら
せて出力する第１のフレームメモリと、前記の第１のフ
レームメモリの出力する前フレームの二値画像と前フレ
ームでの目標候補を指定する領域番号を受けて第１のフ
レームメモリの出力する二値画像の内目標候補領域番号
で指定された領域の画素のみ有意とし、他の画素は有意
でないと判断する目標領域画素判定手段と、前記の目標
領域画素判定手段の出力する二値画像を記憶する第２の
フレームメモリと、目標候補領域を選択する方法として
今回の二値画像と第２のフレームメモリの出力する前回
の目標領域の二値画像の各領域で一致する画素数を計測
し、領域面積に対する一致画素数の割合が許容値以上で
ある領域を目標候補領域とし、その領域番号を出力する
領域画素比較による目標候補選択手段、領域画素比較に
よる目標候補選択手段の出力する複数の目標候補領域の
面積と重心から各目標候補領域の合成重心位置を追尾位
置として出力する追尾位置計測手段を設けたものであ
る。According to the present invention, the area-by-area area measuring means for measuring the area of each area of the binary image, the area-by-area center-of-gravity measuring means for measuring the center of gravity of each area of the binary image, and the binary image delayed by one screen. A first frame memory to be output, a binary image of the previous frame output from the first frame memory, and a binary value to be output from the first frame memory by receiving a region number designating a target candidate in the previous frame A target area pixel determination unit that determines that only pixels in an area designated by the target candidate area number in the image are significant and other pixels are not significant, and a binary image output from the target area pixel determination unit is stored. As a method of selecting the second frame memory and the target candidate area, the number of matching pixels in each area of the binary image of this time and the previous binary image of the target area output from the second frame memory is measured, Vs area A region in which the ratio of the number of matching pixels is equal to or more than an allowable value is set as a target candidate region, and a target candidate selection unit by region pixel comparison that outputs the region number, and a plurality of target candidate regions output by the target candidate selection unit by region pixel comparison The tracking position measuring means for outputting the combined center of gravity position of each target candidate area as the tracking position from the area and the center of gravity of the target position is provided.
【００１１】またこの発明では、二値画像の各領域の面
積を計測する領域毎面積計測手段、二値画像の各領域の
重心を計測する領域毎重心計測手段、二値画像の各領域
の主軸を計測する領域毎主軸計測手段、二値画像を１画
面分遅らせて出力するフレームメモリと、前記のフレー
ムメモリの出力する前フレームの二値画像と前フレーム
での目標候補を指定する領域番号とから二値画像中の領
域番号に対応する全ての領域について主軸を計測する追
尾領域主軸計測手段と、領域毎重心計測手段の出力と追
尾領域主軸計測手段の出力を受けて領域毎の重心と前回
の追尾領域の主軸との距離を計測する重心−主軸間距離
計測手段、領域毎主軸計測手段の出力と追尾領域主軸計
測手段の出力を受けて領域毎の主軸と前回の追尾領域の
主軸のなす角を計測する主軸傾き誤差計測手段、目標を
同定する手段として、重心−主軸間距離計測手段の出力
と主軸傾き誤差計測手段の出力を受けて重心−主軸間距
離と主軸方向誤差がそれぞれ許容値以内である領域すべ
てを目標候補とし、その領域番号を出力する目標候補選
択手段、目標候補選択手段の出力する複数の目標候補領
域の面積と重心から各目標候補領域の合成重心位置を仮
重心として出力する仮重心計測手段、追尾領域の主軸に
対する領域毎の最遠点から追尾領域の主軸及び前記の仮
重心計測手段の出力する仮重心への距離を前回の各距離
と比較し誤差が許容範囲内の領域を目標候補とし、その
領域番号を出力する最遠点−主軸間距離及び最遠点−重
心間距離評価による目標候補選択手段、目標候補選択手
段と前記の最遠点−主軸間距離及び最遠点−重心間距離
評価による目標候補選択手段の出力する複数の目標候補
領域の面積と重心から各目標候補領域の合成重心位置を
追尾位置として出力する追尾位置計測手段を設けたもの
である。Further, according to the present invention, the area-by-area area measuring means for measuring the area of each area of the binary image, the area-by-area centroid measuring means for measuring the gravity center of each area of the binary image, and the main axis of each area of the binary image. Main axis measuring means for each area, a frame memory that outputs a binary image with one screen delayed, and a binary image of the previous frame output by the frame memory and an area number that specifies a target candidate in the previous frame. From the tracking area main axis measuring means for measuring the main axis for all areas corresponding to the area number in the binary image, the output of the center of gravity measuring means for each area and the output of the tracking area main axis measuring means, and the center of gravity of each area and the previous time. Center of gravity-main axis distance measuring means for measuring the distance from the main axis of the tracking area, the output of the main area measuring means for each area and the output of the tracking area main axis measuring means, and the main axis for each area and the main axis of the previous tracking area Total angle As the spindle inclination error measuring means and the means for identifying the target, the center of gravity-spindle distance and the spindle direction error are within the respective allowable values in response to the outputs of the centroid-spindle distance measuring means and the spindle tilt error measuring means. A target candidate selecting unit that outputs all the regions as a target candidate, and a temporary barycenter that outputs the combined barycentric position of each target candidate region from the area and barycenter of the plurality of target candidate regions output by the target candidate selecting unit Center of gravity measuring means, the distance from the farthest point of each area to the main axis of the tracking area to the main axis of the tracking area and the temporary center of gravity output by the temporary center of gravity measuring means is compared with the previous distances and the error is within the allowable range. Is the target candidate, and outputs the area number of the farthest point-main axis distance and the farthest point-target point candidate selecting means by distance evaluation of the center of gravity, the target candidate selecting means and the farthest point-main axis distance, and Far point - in which the combined center of gravity position of each target candidate area from the area and the center of gravity of a plurality of target candidate region outputted from the target candidate selecting means according to the distance between the centers of gravity evaluation provided tracking position measuring means for outputting a tracking position.
【００１２】この発明は、撮像機で得られた濃淡画像を
記憶する第３のフレームメモリ、二値画像の各領域の面
積を計測する領域毎面積計測手段、二値画像の各領域の
重心を計測する領域毎重心計測手段、二値画像を１画面
分遅らせて出力する第１のフレームメモリ、前記の第１
のフレームメモリの出力する前フレームの二値画像と前
フレームでの目標候補を指定する領域番号を受けて第１
のフレームメモリの出力する二値画像の内目標候補領域
番号で指定された領域の画素のみ有意とし、他の画素は
有意でないと判断する目標領域画素判定手段、前記の目
標領域画素判定手段の出力する二値画像を記憶する第２
のフレームメモリと、目標候補領域を選択する方法とし
て今回の二値画像と第２のフレームメモリの出力する前
回の目標領域の二値画像の各領域で一致する画素数を計
測し、領域面積に対する一致画素数の割合が許容値以上
である領域に対応する第３の濃淡画像中の領域の前回と
今回の輝度平均値の誤差が許容範囲内の領域を目標候補
とし、その領域番号を出力する領域画素比較による目標
候補選択手段、領域画素比較による目標候補選択手段の
出力する複数の目標候補領域の面積と重心から各目標候
補領域の合成重心位置を追尾位置として出力する追尾位
置計測手段を設けたものである。According to the present invention, there is provided a third frame memory for storing a grayscale image obtained by an image pickup device, an area-by-area area measuring means for measuring an area of each area of a binary image, and a center of gravity of each area of a binary image. Center-of-gravity measuring means for each region to be measured, first frame memory that outputs a binary image with a delay of one screen, and the above-mentioned first
Receiving the binary image of the previous frame output from the frame memory and the area number designating the target candidate in the previous frame,
Of the binary image output from the frame memory, only the pixels in the region designated by the target candidate region number are significant, and the other pixels are not significant, the target region pixel determining means, and the output of the target region pixel determining means Second storing a binary image
As a method of selecting the target memory area and the target memory area, the number of matching pixels in each area of the current binary image and the previous target area binary image output from the second frame memory is measured, and A region in which the difference between the previous and current luminance average values of the region in the third grayscale image corresponding to a region in which the ratio of the number of matching pixels is equal to or more than the allowable value is within the allowable range is set as a target candidate, and the region number is output. A target candidate selecting means for comparing the area pixels and a tracking position measuring means for outputting a combined center of gravity position of each target candidate area as a tracking position from the areas and the centers of gravity of the plurality of target candidate areas output by the target candidate selecting means by comparing the area pixels are provided. It is a thing.
【００１３】[0013]
【作用】艦船では、甲板と構造物の陰などで、撮像機で
撮像された画像上における輝度差が顕著であり、二値画
像に変換すると領域が分割することがある。艦船の場合
とくに艦橋の前後で分割することが多い。このとき分割
した各領域の主軸方向は、分割前の艦首から艦尾への主
軸方向とほぼ同じと考えられる。また分割した各領域の
重心位置は、分割前の主軸となす距離が小さいと考えら
れる。これより、分割前後の主軸を計測し、分割前の目
標の主軸と分割後の領域の主軸のなす角があるしきい値
以内であり、分割前の目標の主軸と領域の重心との距離
があるしきい値以内であるかどうか判定する。この２つ
の条件を満たす全ての領域を合成した領域を分割前の領
域と同一と判断し、この合成領域の重心点を追尾点とす
ることで追尾可能となる。In a ship, the brightness difference on the image taken by the image pickup device is remarkable due to the shadow of the deck and the structure, and the region may be divided when converted into a binary image. In the case of ships, it is often split before and after the bridge. It is considered that the main axis direction of each of the divided regions at this time is almost the same as the main axis direction from the bow to the stern before the division. It is considered that the distance between the center of gravity of each divided area and the main axis before division is small. From this, the main axis before and after division is measured, and the angle between the main axis of the target before division and the main axis of the area after division is within a certain threshold, and the distance between the main axis of the target before division and the center of gravity of the area is Determine if it is within a certain threshold. It is possible to perform tracking by determining that the area obtained by combining all areas satisfying these two conditions is the same as the area before division, and setting the center of gravity of this combined area as the tracking point.
【００１４】また、分割後の各領域の重心位置が分割前
の領域内に存在する位置であれば、その領域は分割前の
領域の一部であるといえる。よって、分割前の二値画像
を保存し、分割後の各領域の重心位置を計測し、その重
心位置が分割前の二値画像において有意画素である領域
を選択し、追尾点を計測することで追尾継続が可能とな
る。If the position of the center of gravity of each area after division exists in the area before division, it can be said that the area is a part of the area before division. Therefore, save the binary image before division, measure the centroid position of each area after division, select the area where the centroid position is a significant pixel in the binary image before division, and measure the tracking point. Tracking can be continued with.
【００１５】分割後の領域で分割前の領域と同一と判断
された領域から追尾点を求める手法として、分割前の目
標領域と分割後の目標領域とで論理和演算を行い、この
結果得られる領域の重心を計測することで追尾位置を得
ることができる。As a method for obtaining a tracking point from the area determined to be the same as the area before the division in the area after the division, a logical sum operation is performed on the target area before the division and the target area after the division, and the result is obtained. The tracking position can be obtained by measuring the center of gravity of the region.
【００１６】分割後の領域が分割前の目標領域と同一か
どうか判定する手法として、分割後の各領域の画素毎に
分割前の目標領域の画素と一致しているかどうか比較
し、分割後の領域面積に対し分割前の目標領域の画素と
一致している画素数の割合があるしきい値以上である領
域を分割前の領域と同一であると判断し、追尾位置を計
測し、追尾継続することができる。As a method of determining whether the area after division is the same as the target area before division, each pixel in each area after division is compared with whether or not it matches the pixel in the target area before division, and after division. It is determined that the area where the ratio of the number of pixels matching the target area pixel before division to the area area is more than a threshold value is the same as the area before division, the tracking position is measured, and tracking continues. can do.
【００１７】目標領域の一部であるが、分割後の領域の
主軸と分割前の目標領域の主軸との傾き誤差が大きい領
域を同一領域と判定する手法として、主軸が分割前の方
向とほぼ同一で重心と分割前の主軸との距離があるしき
い値以内の領域を目標候補と判定しその合成領域の重心
を計測し、これを仮重心とする。さらに仮重心と前回の
追尾領域の主軸に対する領域内の最遠点との距離と分割
前の最遠点−重心間距離との誤差が許容値以内であり、
前回追尾領域の主軸とこれに対する領域内の最遠点との
距離と分割前の最遠点−主軸間距離との誤差が許容値以
内である領域も分割前の領域と同一であると判断し、追
尾位置を計測し追尾継続することができる。As a method of determining a region having a large inclination error between the main axis of the region after division and the main axis of the target region before division, which is a part of the target region, as a method of determining the same region, the main axis is almost in the direction before division. A region in which the distance between the center of gravity and the main axis before division is within a certain threshold value is determined as a target candidate, the center of gravity of the combined region is measured, and this is set as the temporary center of gravity. Furthermore, the error between the distance between the tentative center of gravity and the farthest point in the area with respect to the main axis of the previous tracking area and the distance between the farthest point before division and the center of gravity is within the allowable value,
The area where the error between the distance between the main axis of the previous tracking area and the farthest point in that area and the distance between the farthest point and the main axis before division is within the allowable value is also judged to be the same as the area before division. , Tracking position can be measured and tracking can be continued.
【００１８】分割後の領域が分割前の目標領域と同一か
どうか判定する手法として、分割後の領域を分割前の目
標領域と画素と一致しているかどうか比較し、分割後の
領域面積に対し分割前の目標領域の画素と一致している
画素数の割合があるしきい値以上である領域で、分割前
後に一致した領域に対応する濃淡画像での領域で分割前
の輝度平均値と分割後の輝度平均値の誤差があるしきい
値以下である領域を分割前の領域と同一であると判断
し、追尾位置を計測し追尾継続することができる。As a method of determining whether the divided area is the same as the undivided target area, the divided area is compared with the undivided target area to see if they match the pixels, and In a region where the ratio of the number of pixels that match the pixels in the target region before division is greater than or equal to a threshold value, in the grayscale image region that corresponds to the regions that match before and after division, divide the average brightness value before division and It is possible to determine that the area in which the error of the subsequent brightness average value is less than a certain threshold value is the same as the area before the division, measure the tracking position, and continue the tracking.
【００１９】[0019]
実施例１．図１はこの発明による装置の一実施例を示す
構成図である。図１において、７は領域毎面積計測手
段、８は領域毎主軸計測手段、９ａはフレームメモリ、
１０は追尾領域主軸計測手段、１１は重心−主軸間距離
計測手段、１２は主軸傾き誤差計測手段、１３は目標候
補選択手段、１４は追尾位置計測手段である。Example 1. FIG. 1 is a block diagram showing an embodiment of the device according to the present invention. In FIG. 1, 7 is an area measuring means for each area, 8 is a main axis measuring means for each area, 9a is a frame memory,
Reference numeral 10 is a tracking area spindle measuring means, 11 is a center of gravity-spindle distance measuring means, 12 is a spindle inclination error measuring means, 13 is a target candidate selecting means, and 14 is a tracking position measuring means.
【００２０】この発明による画像追尾装置は、撮像機１
により濃淡画像を得る。二値化手段２は前記濃淡画像を
二値化し、領域毎面積計測手段７は得られた二値画像の
有意画素で構成される連結領域毎の画素数を計測しこれ
を領域の面積とする。領域毎重心計測手段３は領域毎に
そのモーメントから重心位置を計測する。領域毎主軸計
測手段８では“数２”によりその領域が伸びている方向
を表わす主軸の傾きａｎを出力する。ここでｎは領域を
識別するための番号である。“数２”においてｍｎ２
０、ｍｎ１１、ｍｎ０２は２次モーメント、ｆｎ（ｘ、
ｙ）は図７に示すように領域ｎの重心から相対座標
（ｘ、ｙ）の画素が有意画素のとき１、有意画素でない
とき０を与える。The image tracking device according to the present invention is an image pickup device 1.
To obtain a grayscale image. The binarizing means 2 binarizes the grayscale image, and the area-by-area area measuring means 7 measures the number of pixels for each connected area constituted by significant pixels of the obtained binary image and sets this as the area of the area. . The centroid measuring means 3 for each area measures the centroid position from the moment for each area. The area-by-area main axis measuring means 8 outputs the inclination an of the main axis indicating the direction in which the area is extended, using "Equation 2". Here, n is a number for identifying the area. Mn2 in "Equation 2"
0, mn11, mn02 are second moments, fn (x,
As shown in FIG. 7, y) gives 1 when the pixel of relative coordinates (x, y) from the center of gravity of the region n is a significant pixel, and 0 when it is not a significant pixel.
【００２１】フレームメモリ９ａでは二値画像を記憶し
ている。追尾領域主軸計測手段１０では目標候補選択手
段１３の出力する目標候補領域番号に対応するフレーム
メモリ９ａの出力する二値画像中の領域すべてを追尾領
域とし主軸計測の対象領域として、領域毎主軸計測手段
８と同様に“数２”によって前回追尾領域の主軸の傾き
ａ０を出力する。また、追尾領域の重心位置（Ｘ０、Ｙ
０）も計測し出力する。主軸の傾きと重心すなわち主軸
直線の通過する点を出力することで、主軸直線を表わす
方程式を一意に決定する。図８は追尾領域主軸計測手段
１０の動作の一例を示す図である。図８では領域２、
３、４が目標候補領域として指定され、フレームメモリ
９ａが記憶する二値画像中の領域２、３、４を一つの領
域とみなして追尾領域主軸計測手段１０で主軸の傾きａ
０、重心（Ｘ０、Ｙ０）を出力し、主軸を表わす直線Ｌ
０を決定している。The frame memory 9a stores binary images. In the tracking area spindle measuring means 10, all areas in the binary image output by the frame memory 9a corresponding to the target candidate area number output by the target candidate selecting means 13 are set as tracking areas, and the main axis measurement is performed for each area. Similar to the means 8, the inclination a0 of the main axis of the previous tracking area is output by "Equation 2". Also, the position of the center of gravity of the tracking area (X0, Y
0) is also measured and output. The equation representing the main axis straight line is uniquely determined by outputting the inclination of the main axis and the center of gravity, that is, the point through which the main axis straight line passes. FIG. 8 is a diagram showing an example of the operation of the tracking area spindle measuring means 10. In FIG. 8, area 2,
3 and 4 are designated as target candidate areas, and areas 2, 3 and 4 in the binary image stored in the frame memory 9a are regarded as one area, and the tracking area main axis measuring means 10 causes the inclination a of the main axis a.
A straight line L that outputs 0 and the center of gravity (X0, Y0) and represents the main axis
0 has been determined.
【００２２】[0022]
【数２】 [Equation 2]
【００２３】図９は目標が３つの領域すなわち領域２、
３、４に分割している場合の二値画像の一例を示してい
る。直線Ｌ０は前回の領域の主軸を表わす直線である。
重心−主軸間距離計測手段１１では“数３”により、領
域ｎの重心と前回の追尾領域の主軸である直線Ｌ０との
距離ｄｎを計測する。“数３”において（Ｘｎ、Ｙｎ）
は領域ｎの重心、（Ｘ０、Ｙ０）は前回追尾領域の重
心、ａ０は前回追尾領域の主軸の傾きである。FIG. 9 shows that the target has three regions, that is, region 2,
An example of a binary image when divided into 3, 4 is shown. The straight line L0 is a straight line representing the main axis of the previous region.
The center-of-gravity-main axis distance measuring means 11 measures the distance dn between the center of gravity of the area n and the straight line L0 which is the main axis of the previous tracking area, using “Equation 3”. In “Equation 3” (Xn, Yn)
Is the center of gravity of the region n, (X0, Y0) is the center of gravity of the previous tracking region, and a0 is the inclination of the main axis of the previous tracking region.
【００２４】[0024]
【数３】 (Equation 3)
【００２５】主軸傾き誤差計測手段１２では領域ｎの主
軸の傾きａｎと前回目標領域の主軸の傾きａ０の誤差ｄ
ａｎを計測する。目標候補選択手段１３では、重心−主
軸間距離計測手段１１の出力するｄｎがしきい値以下で
あり、主軸傾き誤差計測手段１２の出力するｄａｎがし
きい値以下である領域を分割前の領域と同一と判断して
目標候補領域とし、その領域番号を出力する。図９では
主軸方向が直線Ｌ０とほぼ同一方向で、重心が直線Ｌ０
に近い領域２、３、４が目標候補領域として選択され
る。追尾位置計測手段１４では目標候補領域の重心（Ｘ
ｉ、Ｙｉ）及び面積Ｓｉから“数４”により追尾位置
（Ｘ、Ｙ）を出力する。これにより目標領域が分割した
場合でも、二値画像の各領域の内目標領域に相当する全
ての領域を一つの領域と見なし、この領域を目標領域と
して追尾継続することができる。In the spindle inclination error measuring means 12, an error d between the inclination an of the spindle of the area n and the inclination a0 of the spindle of the previous target area is d.
Measure an. In the target candidate selecting means 13, the area before the division in which the dn output from the center-of-spindle distance measuring means 11 is less than or equal to the threshold value and the dan output from the main spindle tilt error measuring means 12 is less than or equal to the threshold value And the target candidate area, and outputs the area number. In FIG. 9, the main axis direction is substantially the same as the straight line L0, and the center of gravity is the straight line L0.
Areas 2, 3 and 4 close to are selected as target candidate areas. In the tracking position measuring means 14, the center of gravity (X
i, Yi) and the area Si, the tracking position (X, Y) is output according to "Equation 4". As a result, even if the target area is divided, it is possible to regard all areas corresponding to the target area in each area of the binary image as one area, and continue tracking as this area.
【００２６】[0026]
【数４】 [Equation 4]
【００２７】実施例２．図２はこの発明による画像追尾
装置の第２の実施例を示す構成図である。図２において
９ｂは第１のフレームメモリ、９ｃは第２のフレームメ
モリ、１５は目標領域画素判定手段、１６は重心位置判
定による目標候補選択手段である。Example 2. FIG. 2 is a block diagram showing a second embodiment of the image tracking device according to the present invention. In FIG. 2, 9b is a first frame memory, 9c is a second frame memory, 15 is a target area pixel determining means, and 16 is a target candidate selecting means based on the gravity center position determination.
【００２８】この発明による画像追尾装置は、撮像機１
により濃淡画像を得る。二値化手段２は前記濃淡画像を
二値化し、領域毎重心計測手段３は得られた二値画像の
有意画素で構成される連結領域毎の重心位置を計測す
る。領域毎面積計測手段７では連結領域毎の面積を計測
する。第１のフレームメモリ９ｂでは前フレームでの二
値画像を記憶している。目標領域画素判定手段１５では
第１のフレームメモリ９ｂの出力する前フレームでの二
値画像の内、前回重心位置判定による目標候補選択手段
１６が出力した目標候補領域番号で指定された領域の画
素のみ有意とし、他の画素は有意でないと判断すること
で、前フレームでの目標領域を抽出する。第２のフレー
ムメモリ９ｃでは前記の目標領域画素判定手段１５の出
力する前フレームでの目標領域の二値画像を記憶する。The image tracking device according to the present invention comprises an image pickup device 1.
To obtain a grayscale image. The binarizing unit 2 binarizes the grayscale image, and the region-by-region centroid measuring unit 3 measures the centroid position for each connected region composed of significant pixels of the obtained binary image. The area-by-area area measuring unit 7 measures the area for each connected area. The first frame memory 9b stores the binary image in the previous frame. In the target area pixel determination means 15, the pixels in the area specified by the target candidate area number output by the target candidate selection means 16 by the previous determination of the center of gravity among the binary images in the previous frame output from the first frame memory 9b. The target area in the previous frame is extracted by determining that only the pixels are significant and the other pixels are not significant. The second frame memory 9c stores the binary image of the target area in the previous frame output from the target area pixel determination means 15.
【００２９】重心位置判定による目標候補選択手段１６
では今回計測した領域ｎの重心（Ｘｎ、Ｙｎ）と同位置
の第２のフレームメモリ９ｃ内の画素値が１かどうか調
べ、画素値が１であれば領域ｎを目標候補とし、領域番
号ｎを出力する。追尾位置計測手段１４では“数４”に
より追尾位置を出力する。これにより追尾位置計測以外
に複雑な演算を行うことなく分割した領域が目標領域で
あるかどうか判定することができる。Target candidate selection means 16 based on the determination of the position of the center of gravity
Then, it is checked whether the pixel value in the second frame memory 9c at the same position as the center of gravity (Xn, Yn) of the region n measured this time is 1, and if the pixel value is 1, the region n is set as the target candidate and the region number n Is output. The tracking position measuring means 14 outputs the tracking position according to "Equation 4". This makes it possible to determine whether or not the divided area is the target area without performing a complicated calculation other than the tracking position measurement.
【００３０】実施例３．図３はこの発明による画像追尾
装置の第３の実施例を示す構成図である。図３において
９ｄは第３のフレームメモリ、１７は領域論理和による
追尾位置計測手段である。本実施例では前回と今回の目
標領域の二値画像の論理和演算を行い追尾位置を計測す
る領域論理和による追尾位置計測手段を備えている。Example 3. FIG. 3 is a block diagram showing a third embodiment of the image tracking device according to the present invention. In FIG. 3, 9d is a third frame memory, and 17 is a tracking position measuring means based on area OR. In the present embodiment, there is provided tracking position measuring means based on the area logical sum for performing the logical sum operation of the binary images of the previous and present target areas to measure the tracking position.
【００３１】図１０は領域の論理和の一例を示した図で
ある。領域論理和による追尾位置計測手段１７では図１
０（ａ）に示す第３のフレームメモリ９ｄに記憶されて
いる前フレームでの目標候補領域の二値画像と図１０
（ｂ）に示す第２のフレームメモリ９ｃに記憶されてい
る目標候補領域の二値画像との画素毎の論理和演算を行
い図１０（ｃ）に示す演算結果画像を得て、論理和演算
結果画像の重心位置を計測し追尾位置として出力する。
本実施例によれば、論理和演算結果画像から重心位置を
計測するため、あらかじめ領域毎の面積を計測する必要
がなく、複数の目標候補領域から追尾位置を計測するこ
とができる。FIG. 10 is a diagram showing an example of the logical sum of areas. The tracking position measuring means 17 based on the area logical sum is shown in FIG.
10 (a) and the binary image of the target candidate area in the previous frame stored in the third frame memory 9d and FIG.
The OR operation is performed for each pixel with the binary image of the target candidate area stored in the second frame memory 9c shown in (b) to obtain the operation result image shown in FIG. The center of gravity of the result image is measured and output as the tracking position.
According to the present embodiment, the barycentric position is measured from the logical sum operation result image, so that it is not necessary to measure the area of each region in advance, and the tracking position can be measured from a plurality of target candidate regions.
【００３２】実施例４．図４はこの発明による画像追尾
装置の第４の実施例を示す構成図である。図４において
１８は領域画素比較による目標候補選択手段である。Example 4. FIG. 4 is a block diagram showing a fourth embodiment of the image tracking device according to the present invention. In FIG. 4, reference numeral 18 designates a target candidate selecting means based on area pixel comparison.
【００３３】この発明による画像追尾装置は、撮像機１
により濃淡画像を得る。二値化手段２は前記濃淡画像を
二値化し、領域毎重心計測手段３は得られた二値画像の
有意画素で構成される連結領域毎の重心位置を計測す
る。領域毎面積計測手段７では連結領域毎の面積を計測
する。第１のフレームメモリ９ｂでは前回の二値画像を
記憶している。目標領域画素判定手段１５では第１のフ
レームメモリ９ｂの出力する二値画像の内目標候補領域
番号で指定された領域の画素のみ有意とし、他の画素は
有意でないと判断する。第２のフレームメモリ９ｃでは
前記の目標領域画素判定手段の出力する前回の目標候補
領域の二値画像を記憶する。The image tracking device according to the present invention comprises an image pickup device 1.
To obtain a grayscale image. The binarizing unit 2 binarizes the grayscale image, and the region-by-region centroid measuring unit 3 measures the centroid position for each connected region composed of significant pixels of the obtained binary image. The area-by-area area measuring unit 7 measures the area for each connected area. The first frame memory 9b stores the previous binary image. The target area pixel determination means 15 determines that only the pixels in the area designated by the target candidate area number in the binary image output from the first frame memory 9b are significant, and the other pixels are not significant. The second frame memory 9c stores the previous binary image of the target candidate area output from the target area pixel determination means.
【００３４】図１１は領域画素比較による目標候補選択
手段１８の動作を説明する図である。領域画素比較によ
る目標候補選択手段１８では今回の二値画像の領域ｎ内
の画素と第２のフレームメモリ９ｃの出力する前回の二
値画像の同一座標の画素を比較し、図１１で斜線で示し
た部分の面積すなわち領域ｎ内の画素が前回の二値画像
と有意画素で一致している数Ｓｎ（Ｍａｔｃｈ）を計測
する。さらに“数５”によって領域ｎの面積Ｓｎに対す
る一致画素数Ｓｎ（Ｍａｔｃｈ）の割合Ｍｎを計測し、
Ｍｎがしきい値以上である領域ｎを目標候補とし、領域
番号ｎを出力する。追尾位置計測手段１４では“数４”
により追尾位置を出力する。本実施例では今回計測され
た連結領域の内前回の目標候補領域に一致している面積
の割合がしきい値以上の領域を目標候補領域と判定し追
尾位置を計測することで継続追尾することができる。FIG. 11 is a diagram for explaining the operation of the target candidate selecting means 18 by comparing the area pixels. The target candidate selection means 18 based on area pixel comparison compares the pixel in the area n of the current binary image with the pixel of the same coordinate of the previous binary image output from the second frame memory 9c, and hatched in FIG. The area Sn of the indicated portion, that is, the number Sn (Match) in which the pixels in the region n match the previous binary image as significant pixels is measured. Further, the ratio Mn of the number of matching pixels Sn (Match) to the area Sn of the region n is measured by “Equation 5”,
The region number n is output with the region n in which Mn is equal to or larger than the threshold value as a target candidate. In the tracking position measuring means 14, "Equation 4"
The tracking position is output by. In the present embodiment, continuous tracking is performed by measuring a tracking position by determining a region in which the ratio of the area matching the previous target candidate region of the connected regions measured this time is a threshold value or more and measuring the tracking position. You can
【００３５】[0035]
【数５】 (Equation 5)
【００３６】実施例５．図５はこの発明による画像追尾
装置の第５の実施例を示す構成図である。図５におい
て、１９は仮重心計測手段、２０は最遠点−主軸間距離
及び最遠点−重心間距離評価による目標候補選択手段で
ある。Example 5. FIG. 5 is a block diagram showing a fifth embodiment of the image tracking apparatus according to the present invention. In FIG. 5, reference numeral 19 is a temporary center of gravity measuring means, and 20 is a target candidate selecting means based on the evaluation of the distance between the farthest point and the main axis and the distance between the farthest point and the center of gravity.
【００３７】図１２（ａ），（ｂ）は本実施例の動作を
説明する模式図であり、図１２（ａ）は分割前の二値画
像である。この発明による画像追尾装置は、実施例１と
同様に目標候補選択手段１３において領域の重心−目標
領域主軸間距離と領域毎の主軸傾き誤差を評価し、図１
２（ｂ）の領域２、３のように主軸方向が同一で主軸に
重心位置が近い領域を目標候補としてその領域番号を出
力する。仮重心計測手段１９では目標候補選択手段１３
の選択した目標候補領域毎の重心（Ｘｉ、Ｙｉ）及び面
積Ｓｉから“数４”により複数の目標候補領域を一つの
領域と見なしたときの重心（図１２（ｂ）の点Ｃ’）位
置を出力する。FIGS. 12A and 12B are schematic diagrams for explaining the operation of this embodiment, and FIG. 12A is a binary image before division. The image tracking device according to the present invention evaluates the distance between the center of gravity of the area and the target area main axis and the main axis inclination error for each area in the target candidate selection means 13 as in the first embodiment.
The region number is output as a target candidate for a region having the same main axis direction and a center of gravity close to the main shaft, such as regions 2 and 3 in 2 (b). In the temporary center of gravity measuring means 19, the target candidate selecting means 13
The center of gravity (Xi, Yi) of each selected target candidate area and the center of gravity when the plurality of target candidate areas are regarded as one area from the area Si by "Equation 4" (point C'in FIG. 12B). Output the position.
【００３８】最遠点−主軸間距離及び最遠点−重心間距
離評価による目標候補選択手段２０では、前回の追尾領
域の主軸に対する領域内の最遠点（図１２（ｂ）の点
Ａ’）を何らかの方法で求め、この最遠点と前回の追尾
領域の主軸と距離（Ａ’−Ｂ’）と、最遠点と仮重心計
測手段の出力する仮重心の距離（Ａ’−Ｃ’）の分割前
後での誤差を評価し、それぞれの誤差がしきい値以内で
ある領域を目標候補として出力する。例えば図１２
（ｂ）の領域１。追尾位置計測手段１４では目標候補領
域の重心及び面積から“数４”により追尾位置を出力す
る。本実施例によれば、図１２（ｂ）の領域１のように
目標領域の一部であるが主軸傾き誤差が大きいため重心
−目標領域主軸間距離と領域毎の主軸傾き誤差による目
標候補選択では目標候補に選択されない領域も目標候補
として選択できる。In the target candidate selecting means 20 based on the evaluation of the distance between the farthest point and the main axis and the distance between the farthest point and the center of gravity, the farthest point in the area with respect to the main axis of the previous tracking area (point A'in FIG. 12B). ) By some method, and the distance (A′−B ′) between the farthest point and the main axis of the previous tracking area, and the distance (A′−C ′) between the farthest point and the temporary center of gravity output by the temporary center of gravity measuring means. ), The error before and after the division is evaluated, and the area in which each error is within the threshold value is output as the target candidate. For example, in FIG.
Area 1 in (b). The tracking position measuring means 14 outputs the tracking position according to "Equation 4" from the center of gravity and the area of the target candidate area. According to the present embodiment, a target candidate is selected based on the distance between the center of gravity and the target region main spindle and the main shaft tilt error for each region because the main shaft tilt error is large as in a region 1 of FIG. 12B. Then, an area that is not selected as a target candidate can be selected as a target candidate.
【００３９】実施例６．図６はこの発明による画像追尾
装置の第６の実施例を示す構成図である。図６において
９ｄは第３のフレームメモリ、２１は領域画素比較及び
輝度平均値による目標候補選択手段である。Example 6. FIG. 6 is a block diagram showing a sixth embodiment of the image tracking device according to the present invention. In FIG. 6, 9d is a third frame memory, and 21 is a target candidate selecting means based on area pixel comparison and brightness average value.
【００４０】この発明による画像追尾装置は、実施例４
と同様に二値化手段２で前記濃淡画像を二値化し、領域
毎重心計測手段３で連結領域毎の重心位置を計測し、領
域毎面積計測手段７で連結領域毎の面積を計測し、第２
のフレームメモリ９ｃでは前回の目標候補領域の二値画
像を記憶する。また第３のフレームメモリでは前回の濃
淡画像を記憶する。The image tracking apparatus according to the present invention is the same as that of the fourth embodiment.
Similarly, the binarization unit 2 binarizes the grayscale image, the region-by-region centroid measuring unit 3 measures the barycentric position for each connected region, and the region-based area measuring unit 7 measures the area for each connected region. Second
The frame memory 9c stores the previous binary image of the target candidate area. In addition, the previous grayscale image is stored in the third frame memory.
【００４１】領域画素比較及び輝度平均値による目標候
補選択手段２１では今回の二値画像の領域ｎ内の画素と
第２のフレームメモリ９ｃの出力する前回の二値画像の
同一座標の画素を比較し、領域ｎ内の画素が前回の二値
画像と有意画素で一致している数Ｓｎ（Ｍａｔｃｈ）を
計測し、“数５”によって領域ｎの面積Ｓｎに対する一
致画素数Ｓｎ（Ｍａｔｃｈ）の割合Ｍｎを計測する。さ
らに、二値画像で一致した画素の割合が高い領域に対応
する今回の濃淡画像と第３のフレームメモリ９ｄの記憶
する前回の濃淡画像の領域の輝度平均値を計測しその誤
差を計測する。Ｍｎがしきい値以上であり、輝度平均値
の誤差がしきい値以内である領域を目標候補とし、追尾
位置計測手段１４で“数４”により追尾位置を出力す
る。具体的には、前後のフレームで領域の存在する位置
がほぼ同じで、その輝度もほぼ同じである領域が目標候
補として選択される。本実施例によれば、前フレームで
の目標領域の二値画像と今回の二値画像が画素毎に一致
する割合が高い領域の内で、さらに領域の濃淡画像の輝
度平均値が近い値の領域を目標候補領域と選択すること
で実施例４よりも厳密な目標選択ができる。The target candidate selecting means 21 based on area pixel comparison and brightness average value compares the pixel in the area n of the current binary image with the pixel of the same coordinate of the previous binary image output from the second frame memory 9c. Then, the number Sn (Match) in which the pixels in the region n match the previous binary image with significant pixels is measured, and the ratio of the number of matching pixels Sn (Match) to the area Sn of the region n is calculated by "Equation 5". Measure Mn. Further, the average luminance value of the areas of the current grayscale image corresponding to the area in which the ratio of the matched pixels in the binary image is high and the previous grayscale image stored in the third frame memory 9d is measured, and the error is measured. An area in which Mn is equal to or larger than the threshold value and an error in the average luminance value is within the threshold value is set as a target candidate, and the tracking position measuring unit 14 outputs the tracking position according to "Equation 4". Specifically, the regions where the positions of the regions are substantially the same in the preceding and following frames and the luminance thereof is also the same are selected as target candidates. According to the present embodiment, in the region in which the binary image of the target region in the previous frame and the binary image of this time have a high rate of matching for each pixel, the brightness average value of the grayscale image of the region is closer. By selecting the region as the target candidate region, the target can be selected more strictly than in the fourth embodiment.
【００４２】[0042]
【発明の効果】以上のようにこの発明によれば、前回追
尾領域と定めた領域の主軸を計測するとともに、領域毎
の重心及び主軸方向を計測し、前回追尾領域の主軸直線
と領域の重心との距離が許容値内であり、主軸方向も誤
差も許容値内である領域を目標候補領域とし、全ての目
標候補領域を合成した領域の重心を追尾位置として出力
する。これにより目標領域が分割した場合でも、分割し
た各領域を目標の一部として扱い、目標を追尾継続する
ことができる。As described above, according to the present invention, the main axis of the area defined as the previous tracking area is measured, and the center of gravity and the main axis direction of each area are measured to determine the main axis straight line of the previous tracking area and the center of gravity of the area. An area in which the distance between and is within the allowable value and both the main axis direction and the error are within the allowable value is set as the target candidate area, and the center of gravity of the area in which all the target candidate areas are combined is output as the tracking position. As a result, even when the target area is divided, each divided area can be treated as a part of the target, and the target can be continuously tracked.
【００４３】またこの発明によれば、領域毎に計測した
重心位置が、記憶している前回の目標領域内に位置する
領域を目標候補領域とすることで、分割した領域が目標
領域であるかどうか判定することができる。Further, according to the present invention, the divided area is the target area by setting the area where the barycentric position measured for each area is within the stored previous target area as the target candidate area. You can judge.
【００４４】さらにこの発明によれば、前回の目標領域
と今回の目標領域を画素毎に論理和演算し、その結果生
じる領域を目標領域としその重心を追尾位置とすること
で、複数の目標候補領域から追尾位置を計測することが
できる。Further, according to the present invention, a plurality of target candidates are obtained by logically ORing the previous target area and the current target area for each pixel, and setting the resulting area as the target area and the center of gravity thereof as the tracking position. The tracking position can be measured from the area.
【００４５】この発明によれば、今回計測された連結領
域の内前回の目標候補領域に一致している面積の割合を
評価し、一致する割合が許容値以上の領域を目標候補領
域と判定し追尾位置を計測することで継続追尾すること
ができる。According to the present invention, the ratio of the area that matches the previous target candidate area in the connected area measured this time is evaluated, and the area where the matching ratio is the allowable value or more is determined as the target candidate area. Continuous tracking can be performed by measuring the tracking position.
【００４６】この発明によれば、前回追尾領域の主軸直
線と領域の重心との距離と、前回追尾領域の主軸と領域
の主軸の傾き誤差を評価し、主軸に対する領域内最遠点
と主軸の距離と、最遠点と仮重心の距離の分割前後での
誤差を評価することで、目標領域の一部であるが主軸傾
き誤差が大きい領域も目標候補領域と判定し継続追尾す
ることができる。According to this invention, the distance between the main axis straight line of the previous tracking area and the center of gravity of the area and the inclination error between the main axis of the previous tracking area and the main axis of the area are evaluated, and the farthest point in the area with respect to the main axis and the main axis. By evaluating the distance and the error before and after the division of the distance between the farthest point and the provisional center of gravity, a region that is a part of the target region but has a large spindle tilt error can be determined as a target candidate region and can be continuously tracked. .
【００４７】またこの発明によれば、今回計測された連
結領域が前回の目標候補領域に一致している割合が許容
値以上の一致度を示し、一致した互いの領域の輝度平均
値の誤差が許容値以内である領域を目標候補領域と判定
し継続追尾することができる。Further, according to the present invention, the rate at which the connected region measured this time matches the previous target candidate region indicates a degree of coincidence equal to or greater than the allowable value, and the error of the average brightness value of the mutually matched regions is reduced. An area that is within the allowable value can be determined as a target candidate area and can be continuously tracked.
【図１】 この発明による画像追尾装置の第１の実施例
を示す構成図である。FIG. 1 is a configuration diagram showing a first embodiment of an image tracking device according to the present invention.
【図２】 この発明による画像追尾装置の第２の実施例
を示す構成図である。FIG. 2 is a configuration diagram showing a second embodiment of the image tracking device according to the present invention.
【図３】 この発明による画像追尾装置の第３の実施例
を示す構成図である。FIG. 3 is a configuration diagram showing a third embodiment of an image tracking device according to the present invention.
【図４】 この発明による画像追尾装置の第４の実施例
を示す構成図である。FIG. 4 is a configuration diagram showing a fourth embodiment of the image tracking device according to the present invention.
【図５】 この発明による画像追尾装置の第５の実施例
を示す構成図である。FIG. 5 is a configuration diagram showing a fifth embodiment of an image tracking device according to the present invention.
【図６】 この発明による画像追尾装置の第６の実施例
を示す構成図である。FIG. 6 is a configuration diagram showing a sixth embodiment of an image tracking device according to the present invention.
【図７】 重心からの相対座標による画素の指定を説明
する図である。FIG. 7 is a diagram illustrating designation of pixels by relative coordinates from the center of gravity.
【図８】 追尾領域主軸計測手段の動作の一例を示す図
である。FIG. 8 is a diagram illustrating an example of an operation of a tracking area spindle measuring unit.
【図９】 目標が３つの領域（領域２、３、４）に分割
している場合の二値画像の一例を示す図である。FIG. 9 is a diagram showing an example of a binary image when a target is divided into three regions (regions 2, 3, 4).
【図１０】 領域の論理和の一例を示した図である。FIG. 10 is a diagram showing an example of a logical sum of areas.
【図１１】 領域画素比較による目標候補選択手段の動
作を説明する図である。FIG. 11 is a diagram for explaining the operation of the target candidate selecting means based on area pixel comparison.
【図１２】 実施例５の動作を説明する模式図である。FIG. 12 is a schematic diagram illustrating the operation of the fifth embodiment.
【図１３】 従来の画像追尾装置を示す構成図である。FIG. 13 is a configuration diagram showing a conventional image tracking device.
【図１４】 目標領域が分割する前後の模式的な二値画
像である。FIG. 14 is a schematic binary image before and after a target area is divided.
７ 領域毎面積計測手段、８ 領域毎主軸計測手段、９
ａ フレームメモリ、９ｂ 第１のフレームメモリ、９
ｃ 第２のフレームメモリ、９ｄ 第３のフレームメモ
リ、１０ 追尾領域主軸計測手段、１１ 重心−主軸間
距離計測手段、１２ 主軸傾き誤差計測手段、１３ 目
標候補選択手段、１４ 追尾位置計測手段、１５ 目標
領域画素判定手段、１６ 重心位置判定による目標候補
選択手段、１７ 領域論理和による追尾位置計測手段、
１８ 領域画素比較による目標候補選択手段、１９ 仮
重心計測手段、２０ 最遠点−主軸間距離及び最遠点−
重心間距離評価による目標候補選択手段、２１ 領域画
素比較及び輝度平均値比較による目標候補選択手段。7 area measuring means for each area, 8 spindle measuring means for each area, 9
a frame memory, 9b first frame memory, 9
c second frame memory, 9d third frame memory, 10 tracking area spindle measuring means, 11 center of gravity-spindle distance measuring means, 12 spindle tilt error measuring means, 13 target candidate selecting means, 14 tracking position measuring means, 15 Target area pixel determining means, 16 target candidate selecting means by barycentric position determination, 17 tracking position measuring means by area OR,
18 Target Candidate Selection Means by Region Pixel Comparison, 19 Temporary Center of Gravity Measurement Means, 20 Farthest Point-Distance between Spindles and Farthest Point-
Target candidate selecting means by distance between centers of gravity, 21 Target candidate selecting means by area pixel comparison and brightness average value comparison.
フロントページの続き (51)Int.Cl.6 識別記号 庁内整理番号 ＦＩ 技術表示箇所 Ｇ０６Ｔ 1/00 7/20 Ｈ０４Ｎ 7/18 Ｇ Continuation of the front page (51) Int.Cl. 6 Identification code Office reference number FI Technical display location G06T 1/00 7/20 H04N 7/18 G
Claims (6)
二値化処理することによって得られた画面中の領域から
有意な領域を目標として選択し、その領域の重心位置を
追尾位置として出力する画像追尾装置において、画像信
号を二値化する二値化手段と、前記二値化手段によって
得られた二値画像の領域毎の面積を計測する領域毎面積
計測手段と、前記二値化手段によって得られた二値画像
の領域毎の重心を計測する領域毎重心計測手段と、前記
二値化手段によって得られた二値画像の領域毎の主軸を
計測する領域毎主軸計測手段と、前記二値化手段の出力
する二値画像を１画面分遅らせて出力するフレームメモ
リと、前記フレームメモリの出力する前フレームの二値
画像と前フレームでの目標候補を指定する領域番号とか
ら二値画像中の領域番号に対応する全ての領域を１つの
領域とみなしこれを前回の追尾領域としてその主軸を計
測する追尾領域主軸計測手段と、前記領域毎重心計測手
段の出力する重心位置と追尾領域主軸計測手段の出力す
る前回の追尾領域の主軸との距離を計測する重心−主軸
間距離計測手段と、前記領域毎主軸計測手段の出力する
主軸と追尾領域主軸計測手段の出力する前回の追尾領域
の主軸のなす角を計測する主軸傾き誤差計測手段と、前
記重心−主軸間距離計測手段の出力と前記主軸傾き誤差
計測手段の出力を受けて、重心−主軸間距離および主軸
の傾き誤差がそれぞれ許容範囲内の領域を今回の目標候
補領域とし、その領域番号を出力する目標候補選択手段
と、前記目標候補選択手段の出力と前記の領域毎面積計
測手段の出力および前記領域毎重心計測手段の出力とか
ら追尾位置を計測する追尾位置計測手段とを備えたこと
を特徴とする画像追尾装置。1. A significant region is selected as a target from a region on a screen obtained by obtaining an image signal from an image pickup device and binarizing the image signal, and a barycentric position of the region is set as a tracking position. In an image tracking device for outputting, binarizing means for binarizing an image signal, area-by-area area measuring means for measuring an area for each area of a binary image obtained by the binarizing means, and the binary Area-by-area centroid measuring means for measuring the centroid of each area of the binary image obtained by the binarizing means, and area-by-axis spindle measuring means for measuring the spindle of each area of the binary image obtained by the binarizing means, A frame memory that outputs the binary image output by the binarizing unit with a delay of one screen, a binary image of the previous frame output by the frame memory, and an area number that specifies a target candidate in the previous frame. Area in binary image All areas corresponding to the numbers are regarded as one area, and this area is used as the previous tracking area to measure its main axis. Tracking area spindle measuring means, and the center of gravity position and tracking area spindle measuring means output by the centroid measuring means for each area. The center of gravity-main spindle distance measuring means for measuring the distance from the spindle of the previous tracking area to be output, and the spindle output from the spindle measuring means for each area and the spindle of the previous tracking area output from the tracking area spindle measuring means. In response to the output of the spindle inclination error measuring means for measuring an angle, the center of gravity-spindle distance measuring means and the output of the spindle inclination error measuring means, the center of gravity-spindle distance and the inclination error of the spindle are each within an allowable range. An area is set as a target candidate area for this time, and target candidate selecting means for outputting the area number, output of the target candidate selecting means, output of the area-by-area area measuring means, and area-by-area centroid Image tracking apparatus being characterized in that a tracking position measuring means for measuring a tracking position from the output means.
二値化処理することによって得られた画面中の領域から
有意な領域を目標として選択し、その領域の重心位置を
追尾位置として出力する画像追尾装置において、画像信
号を二値化する二値化手段と、前記二値化手段によって
得られた二値画像の領域毎の重心を計測する領域毎重心
計測手段と、前記二値化手段によって得られた二値画像
の領域毎の面積を計測する領域毎面積計測手段と、前記
二値化手段の出力する二値画像を１画面分遅らせて出力
する第１のフレームメモリと、前記第１のフレームメモ
リの出力と前フレームでの目標候補を指定する領域番号
を受けて前記第１のフレームメモリの出力する二値画像
の内目標候補領域番号で指定された領域の画素のみ有意
とし、他の画素は有意でないと判断する目標領域画素判
定手段と、前記目標領域画素判定手段の出力する二値画
像を記憶する第２のフレームメモリと、前記領域毎重心
計測手段の出力する各領域の重心位置が前記第２のフレ
ームメモリの二値画像において有意画素であれば目標候
補領域とし、その領域番号を出力する重心位置判定によ
る目標候補選択手段と、前記重心位置判定による目標候
補選択手段の出力と前記領域毎重心計測手段の出力およ
び前記領域毎面積計測手段の出力から追尾位置を計測す
る追尾位置計測手段とを備えたことを特徴とする画像追
尾装置。2. A significant region is selected as a target from a region on the screen obtained by obtaining an image signal from an image pickup device and binarizing the image signal, and the barycentric position of the region is set as a tracking position. In an image tracking device for outputting, binarizing means for binarizing an image signal, area centroid measuring means for measuring the centroid of each area of the binary image obtained by the binarizing means, and the binary Area-by-area area measuring means for measuring the area of each area of the binary image obtained by the binarizing means, and a first frame memory for delaying and outputting the binary image output by the binarizing means by one screen, Receiving the output of the first frame memory and the area number designating the target candidate in the previous frame, only the pixels in the area designated by the target candidate area number of the binary image output by the first frame memory are significant. And other pixels are The target area pixel determining means for determining that it is not appropriate, the second frame memory for storing the binary image output by the target area pixel determining means, and the center-of-gravity position of each area output by the area-by-area center-of-gravity measuring means are If a significant pixel in the binary image of the second frame memory is a target candidate area, the target candidate selecting means for determining the center of gravity position for outputting the area number, the output of the target candidate selecting means for determining the center of gravity position, and the area An image tracking apparatus comprising: a tracking position measuring unit that measures a tracking position from the output of each centroid measuring unit and the output of each area measuring unit.
二値化処理することによって得られた画面中の領域から
有意な領域を目標として選択し、その領域の重心位置を
追尾位置として出力する画像追尾装置において、画像信
号を二値化する二値化手段と、前記二値化手段によって
得られた二値画像の領域毎の重心を計測する領域毎重心
計測手段と、前記二値化手段によって得られた二値画像
の領域毎の主軸を計測する領域毎主軸計測手段と、前記
二値化手段の出力する二値画像を１画面分遅らせて出力
する第１のフレームメモリと、前記第１のフレームメモ
リの出力する前フレームの二値画像と前フレームでの目
標候補を指定する領域番号を受けて二値画像中の領域番
号に対応する全ての領域を１つの領域とみなしこれを前
回の追尾領域としてその主軸を計測する追尾領域主軸計
測手段と、前記領域毎重心計測手段の出力する重心位置
と追尾領域主軸計測手段の出力する前回の追尾領域の主
軸との距離を計測する重心−主軸間距離計測手段と、前
記領域毎主軸計測手段の出力する主軸と追尾領域主軸計
測手段の出力する前回の追尾領域の主軸のなす角を計測
する主軸傾き誤差計測手段と、前記重心−主軸間距離計
測手段の出力と前記主軸傾き誤差計測手段の出力が許容
範囲内かどうか判定し、許容範囲内の領域を目標候補と
し、その領域番号を出力する目標候補選択手段と、前記
第１のフレームメモリの出力と目標候補を指定する領域
番号を受けて第１のフレームメモリの出力する二値画像
の内目標候補領域番号で指定された領域の画素のみ有意
とし、他の画素は有意でないと判断する目標領域画素判
定手段と、前記目標領域画素判定手段の出力する二値画
像を記憶する第２のフレームメモリと、前記第２のフレ
ームメモリの出力を受けて前回の目標候補領域の二値画
像を記憶する第３のフレームメモリと、前記第２のフレ
ームメモリの出力する今回の目標候補領域の二値画像と
前記の第３のフレームメモリの出力する前回の目標領域
の二値画像を論理和した領域の重心位置を追尾位置とし
て出力する領域論理和による追尾位置計測手段とを備え
たことを特徴とする画像追尾装置。3. A significant area is selected as a target from an area on the screen obtained by obtaining an image signal from an image pickup device and binarizing the image signal, and the barycentric position of the area is set as a tracking position. In an image tracking device for outputting, binarizing means for binarizing an image signal, area centroid measuring means for measuring the centroid of each area of the binary image obtained by the binarizing means, and the binary An area-by-area spindle measuring means for measuring a spindle for each area of the binary image obtained by the binarizing means, and a first frame memory for delaying and outputting the binary image output by the binarizing means by one screen, The binary image of the previous frame output from the first frame memory and the area number designating the target candidate in the previous frame are received, and all areas corresponding to the area numbers in the binary image are regarded as one area. As the previous tracking area A tracking area spindle measuring means for measuring the spindle, and a centroid-spindle distance measurement for measuring the distance between the centroid position output by the centroid measuring means for each area and the spindle of the previous tracking area output by the tracking area spindle measuring means. Means for measuring the angle between the spindle output from the spindle measuring means for each area and the spindle of the previous tracking area output from the tracking area spindle measuring means, and the center of gravity-spindle distance measuring means. A target candidate selecting unit that determines whether the output and the output of the spindle inclination error measuring unit are within an allowable range, sets a region within the allowable range as a target candidate, and outputs the region number, and an output of the first frame memory. Receiving the area number designating the target candidate, it is determined that only the pixels in the area designated by the target candidate area number in the binary image output from the first frame memory are significant and the other pixels are not significant. Target area pixel determination means, a second frame memory that stores the binary image output by the target area pixel determination means, and a previous binary image of the target candidate area that is output by the second frame memory. The third frame memory to be stored, the binary image of the current target candidate area output from the second frame memory, and the binary image of the previous target area output from the third frame memory are logically ORed. An image tracking device, comprising: a tracking position measuring means based on a logical sum of areas for outputting the position of the center of gravity of the area as a tracking position.
二値化処理することによって得られた画面中の領域から
有意な領域を目標として選択し、その領域の重心位置を
追尾位置として出力する画像追尾装置において、画像信
号を二値化する二値化手段と、前記二値化手段によって
得られた二値画像の領域毎の重心を計測する領域毎重心
計測手段と、前記二値化手段によって得られた二値画像
の領域毎の面積を計測する領域毎面積計測手段と、前記
二値化手段の出力する二値画像を１画面分遅らせて出力
する第１のフレームメモリと、前記第１のフレームメモ
リの出力と前フレームでの目標候補を指定する領域番号
を受けて第１のフレームメモリの出力する二値画像の内
目標候補領域番号で指定された領域の画素のみ有意と
し、他の画素は有意でないと判断する目標領域画素判定
手段と、前記目標領域画素判定手段の出力する二値画像
を記憶する第２のフレームメモリと、前記二値化手段の
出力する二値画像と前記第２のフレームメモリの出力す
る二値画像の領域毎に一致する画素数を計測し、領域面
積に対する一致画素の割合が許容値以上である領域を目
標候補とし、その領域番号を出力する領域画素比較によ
る目標候補選択手段と、前記領域画素比較による目標候
補選択手段の出力と前記領域毎重心計測手段の出力およ
び前記領域毎面積計測手段の出力から追尾位置を計測す
る追尾位置計測手段とを備えたことを特徴とする画像追
尾装置。4. An image signal is obtained from an image pickup device, and a significant region is selected as a target from the region on the screen obtained by binarizing the image signal, and the barycentric position of the region is set as a tracking position. In an image tracking device for outputting, binarizing means for binarizing an image signal, area centroid measuring means for measuring the centroid of each area of the binary image obtained by the binarizing means, and the binary Area-by-area area measuring means for measuring the area of each area of the binary image obtained by the binarizing means, and a first frame memory for delaying and outputting the binary image output by the binarizing means by one screen, Receiving the output of the first frame memory and the area number designating the target candidate in the previous frame, only the pixels in the area designated by the target candidate area number of the binary image output by the first frame memory are significant. , Other pixels are significant Target area pixel determining means for determining that there is no, a second frame memory for storing the binary image output by the target area pixel determining means, a binary image output by the binarizing means and the second frame The number of matching pixels is measured for each area of the binary image output from the memory, and the area where the ratio of matching pixels to the area area is equal to or greater than the allowable value is set as the target candidate. A selection unit, and a tracking position measurement unit that measures a tracking position from the output of the target candidate selection unit based on the region pixel comparison, the output of the region-by-region centroid measurement unit, and the region-by-region area measurement unit. Image tracking device.
測手段の出力と領域毎重心計測手段の出力から重心を計
測する仮重心計測手段と、二値化手段によって得られた
二値画像と前記の仮重心計測手段の出力と追尾領域主軸
計測手段の出力を受けて追尾領域の主軸に対する領域毎
の最遠点から追尾領域の主軸及び前記の仮重心計測手段
の出力する仮重心への距離を前回の各距離と比較してそ
の誤差が許容範囲内にある領域を目標候補とし、その領
域番号を出力する最遠点−主軸間距離及び最遠点−重心
間距離評価による目標候補選択手段と、目標候補選択手
段の出力と前記最遠点−主軸間距離及び最遠点−重心間
距離評価による目標候補選択手段の出力と領域毎面積計
測手段の出力と領域毎重心計測手段の出力から追尾位置
を計測する追尾位置計測手段とを備えたことを特徴とす
る請求項１記載の画像追尾装置。5. A temporary centroid measuring means for measuring a centroid from the output of the target candidate selecting means, the output of the area-by-area area measuring means, and the output of the area-by-area centroid measuring means, and a binary image obtained by the binarizing means. The distance from the furthest point of each area with respect to the main axis of the tracking area to the main axis of the tracking area and the temporary center of gravity output by the temporary center of gravity measuring means in response to the output of the temporary center of gravity measuring means and the output of the tracking area main axis measuring means. Is selected as a target candidate, and the region having the error within the allowable range is output as the target candidate, and the farthest point-spindle distance and the farthest point-center of gravity distance evaluation target candidate selection means. From the output of the target candidate selection means and the output of the target candidate selection means, the output of the area-by-area area measurement means, and the output of the area-by-area centroid measurement means by the farthest point-spindle distance and the farthest point-centroid distance evaluation. Tracking position to measure tracking position The image tracking device according to claim 1, further comprising a measuring unit.
３のフレームメモリと、二値化手段によって得られた二
値画像と第２のフレームメモリの出力と撮像機で得られ
た濃淡画像と前記の第３のフレームメモリの出力を受け
て二値画像の領域毎に一致する画素数を計測し、領域面
積に対する一致画素数の割合が許容値以上の領域に対応
する濃淡画像中の領域の前回と今回の輝度平均値の誤差
が許容範囲内にある領域を目標候補とし、その領域番号
を出力する領域画素比較及び輝度平均値比較による目標
候補選択手段と、前記領域画素比較及び輝度平均値比較
による目標候補選択手段の出力と領域毎重心計測手段の
出力と領域毎面積計測手段の出力から追尾位置を計測す
る追尾位置計測手段とを備えたことを特徴とする請求項
４記載の画像追尾装置。6. A third frame memory for storing a grayscale image obtained by the image pickup device, a binary image obtained by the binarizing means, an output of the second frame memory, and a grayscale image obtained by the image pickup device. The number of matching pixels for each area of the binary image is measured by receiving the image and the output of the third frame memory, and the grayscale image in the grayscale image corresponding to the area in which the ratio of the number of matching pixels to the area area is equal to or greater than the allowable value. A target candidate selecting means by area pixel comparison and brightness average value comparison, which outputs a region number, in which an area in which the difference between the previous and present brightness average values is within an allowable range is set as a target candidate, and the area pixel comparison and brightness The tracking position measuring means for measuring the tracking position from the output of the target candidate selecting means, the output of the area-by-area centroid measuring means, and the output of the area-by-area area measuring means by means of average value comparison is provided. Image tracking Place.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP6147880A JPH0815401A (en) | 1994-06-29 | 1994-06-29 | Image tracking apparatus |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP6147880A JPH0815401A (en) | 1994-06-29 | 1994-06-29 | Image tracking apparatus |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| JPH0815401A true JPH0815401A (en) | 1996-01-19 |
Family
ID=15440315
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP6147880A Pending JPH0815401A (en) | 1994-06-29 | 1994-06-29 | Image tracking apparatus |
Country Status (1)
| Country | Link |
|---|---|
| JP (1) | JPH0815401A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2014021531A (en) * | 2012-07-12 | 2014-02-03 | Dwango Co Ltd | Program, image processing method, and image processor |
| DE102014213556B4 (en) * | 2013-07-11 | 2015-09-17 | Panasonic Corporation | Tracking support device, tracking support system and tracking support method |
| JP2018503200A (en) * | 2014-11-24 | 2018-02-01 | レイセオン カンパニー | Method and system for processing a series of images to identify, track and / or target objects in a water body |
-
1994
- 1994-06-29 JP JP6147880A patent/JPH0815401A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2014021531A (en) * | 2012-07-12 | 2014-02-03 | Dwango Co Ltd | Program, image processing method, and image processor |
| DE102014213556B4 (en) * | 2013-07-11 | 2015-09-17 | Panasonic Corporation | Tracking support device, tracking support system and tracking support method |
| US9251599B2 (en) | 2013-07-11 | 2016-02-02 | Panasonic Intellectual Property Management Co., Ltd. | Tracking assistance device, a tracking assistance system and a tracking assistance method that enable a monitoring person to perform a task of correcting tracking information |
| JP2018503200A (en) * | 2014-11-24 | 2018-02-01 | レイセオン カンパニー | Method and system for processing a series of images to identify, track and / or target objects in a water body |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US6002793A (en) | Machine vision method and apparatus for finding an object orientation angle of a rectilinear object | |
| US7024041B2 (en) | Pattern inspection apparatus and method | |
| US7474354B2 (en) | Image angle detection device and scan line interpolation device having the same | |
| JP2760250B2 (en) | Pinhole inspection equipment | |
| JPH0815401A (en) | Image tracking apparatus | |
| JP2000003436A (en) | Device and method for recognizing isar picture | |
| US5570298A (en) | Dot pattern-examining apparatus | |
| JP3127598B2 (en) | Method for extracting density-varying constituent pixels in image and method for determining density-fluctuation block | |
| JP3078685B2 (en) | Location recognition method | |
| JP3461143B2 (en) | Color image target position detection device | |
| JP3482265B2 (en) | Steel inspection equipment | |
| JP3585977B2 (en) | Moving area detection device | |
| JPH10213639A (en) | Target tracking device | |
| JP3024231B2 (en) | Photometric calculation device | |
| JP2738252B2 (en) | Edge detection device of strip material by image | |
| JP3240788B2 (en) | Image signal binary decision circuit | |
| JP3050283B2 (en) | Moving object recognition device | |
| JPH04203898A (en) | Image follow-up device | |
| JPH08167023A (en) | Road white line detecting device | |
| JPS61187637A (en) | Apparatus for inspecting appearance | |
| JPH06174818A (en) | Image tracking device | |
| JPH01113603A (en) | Position detecting apparatus | |
| JPH09331522A (en) | Binarization processing unit | |
| JPH04310853A (en) | Device for detecting flatness defect | |
| JP2005309795A (en) | Same object determination device, same object determination method, computer-readable recording medium to which program is recorded, and program |