公开(公告)号：US20100103259A1

公开(公告)日：2010-04-29

申请号：US12582501

申请日：2009-10-20

申请人： Jun TANIDA ,  Ryoichi Horisaki ,  Yoshizumi Nakao ,  Takashi Toyoda ,  Yasuo Masaki

发明人： Jun TANIDA ,  Ryoichi Horisaki ,  Yoshizumi Nakao ,  Takashi Toyoda ,  Yasuo Masaki

IPC分类号： H04N9/47

CPC分类号： G01C3/08 ,  G01B11/026 ,  G02B3/0043 ,  G02B7/38 ,  G02B27/58

摘要： An object distance deriving device comprises: a compound-eye imaging device having imaging units with optical lenses randomly arranged for the respective imaging units; and a distance calculation unit to calculate an object distance using images captured by the compound-eye imaging device. The distance calculation unit: sets temporary distances z (S1); calculates an imaging process matrix [Hz] according to a temporary distance z (S2); estimates a high-resolution image by super-resolution processing using the imaging process matrix [Hz] (S3); uses the estimated high-resolution image to calculate an evaluation value distribution E for evaluating the temporary distance z (S4); repeats steps S2 to S4 for all temporary distances z (S5); and determines, as an object distance, one temporary distance z giving a minimum evaluation value in the evaluation value distributions E. This makes it possible to accurately derive the object distance even if the baseline length of the compound-eye imaging device is limited.

摘要翻译： 物体距离导出装置包括：具有成像单元的复眼成像装置，其具有随机地布置用于各个成像单元的光学透镜; 以及距离计算单元，使用由复眼成像装置拍摄的图像来计算对象距离。 距离计算单元：设定临时距离z（S1）; 根据临时距离z（S2）计算成像过程矩阵[Hz]; 使用成像处理矩阵[Hz]（S3）通过超分辨率处理估计高分辨率图像; 使用估计的高分辨率图像来计算用于评估临时距离z的评估值分布E（S4）; 对于所有临时距离z重复步骤S2至S4（S5）; 并且将评估值分布E中给出最小评估值的一个临时距离z确定为对象距离。这使得即使限制复眼目标成像装置的基线长度，也可以准确地导出对象距离。

公开(公告)号：US20060072029A1

公开(公告)日：2006-04-06

申请号：US10532659

申请日：2003-10-24

申请人： Shigehiro Miyatake ,  Jun Tanida ,  Kenji Yamada

发明人： Shigehiro Miyatake ,  Jun Tanida ,  Kenji Yamada

IPC分类号： H04N5/225 ,  H01L27/00

CPC分类号： G02B3/0056 ,  H04N5/2253 ,  H04N5/2254 ,  H04N5/23229 ,  H04N9/045

摘要： An image input apparatus which reconfigures a single reconfigured image from a plurality of low-resolution, object reduced images formed in a specified region on the light detecting element by the micro-lens array, wherein a high-resolution, single reconfigured image can be obtained even if the distance between the subject and the micro-lens array is long (infinitely long, for example), and further a reconfigured image can be realized in colors. The image input apparatus is characterized in that the relative distance between a micro-lens (1a) and light detecting cells (3a) in a specified region, where object reduced images corresponding to the micro-lens (1a) are formed, is different in each micro-lenses (1a). In addition, the light detecting cells (3a) are divided into a plurality of regions, and color filters (primary color filter, or complementary color filter, for example) are disposed in each of the divided regions.

摘要翻译： 一种图像输入装置，其通过微透镜阵列从形成在光检测元件上的指定区域中的多个低分辨率对象缩小图像重新配置单个重新配置的图像，其中可以获得高分辨率单个重新配置的图像 即使被摄体和微透镜阵列之间的距离长（例如无限长），并且还可以以颜色实现重新配置的图像。 图像输入装置的特征在于，在形成有对应于微透镜（1a）的物体缩小图像的特定区域中的微透镜（1a）和光检测单元（3a）之间的相对距离， 在每个微透镜（1a）中是不同的。 此外，光检测单元（3a）被分成多个区域，并且在每个分割区域中设置滤色器（例如，原色滤色器或补色滤色器）。

公开(公告)号：US20100053600A1

公开(公告)日：2010-03-04

申请号：US12551088

申请日：2009-08-31

申请人： Jun Tanida ,  Ryoichi Horisaki ,  Yoshizumi Nakao ,  Takashi Toyoda ,  Yasuo Masaki

发明人： Jun Tanida ,  Ryoichi Horisaki ,  Yoshizumi Nakao ,  Takashi Toyoda ,  Yasuo Masaki

IPC分类号： G01B9/00

CPC分类号： H04N5/2259 ,  H01L27/14625 ,  H01L27/14685 ,  H04N5/2254 ,  H04N5/349

摘要： An imaginary object plane is set in front of an imaging device body (plane setting step). A part of optical conditions of optical lenses are changed as variables, and positions of points (pixel observation points) on the imaginary object plane where lights coming from pixels of a solid-state imaging element and back-projected through the optical lenses are calculated (pixel observation point calculating step). The dispersion in position of the calculated pixel observation points is evaluated (evaluating step). Finally, a set of values of the variables giving maximum evaluated dispersion of the calculated pixel observation points is determined as optimum optical condition of the optical lenses (condition determining step). This reduces the number of pixels which image the same portions of the target object, making it possible to reduce portions of the same image information in multiple unit images, and to stably obtain a reconstructed image having a high definition.

摘要翻译： 将虚拟物平面设置在成像装置主体的前方（平面设置步骤）。 光学透镜的光学条件的一部分作为变量而变化，并且在计算出来自固态成像元件的像素并通过光学透镜反投影的光的虚拟物面上的点（像素观察点）的位置（ 像素观察点计算步骤）。 评估计算出的像素观察点的位置偏差（评价步骤）。 最后，将给出计算出的像素观察点的最大评估色散的变量的值的集合确定为光学透镜的最佳光学条件（条件确定步骤）。 这减少了对目标对象的相同部分进行成像的像素数量，使得可以减少多个单位图像中的相同图像信息的部分，并且稳定地获得具有高清晰度的重建图像。

公开(公告)号：US07009652B1

公开(公告)日：2006-03-07

申请号：US09639990

申请日：2000-08-16

申请人： Jun Tanida ,  Kenji Yamada ,  Daisuke Miyazaki ,  Yoshiki Ichioka ,  Shigehiro Miyatake ,  Kouichi Ishida

发明人： Jun Tanida ,  Kenji Yamada ,  Daisuke Miyazaki ,  Yoshiki Ichioka ,  Shigehiro Miyatake ,  Kouichi Ishida

IPC分类号： H04N5/225

CPC分类号： H01L27/1462 ,  G02B3/0018 ,  G02B3/0056 ,  H01L27/14627 ,  H01L27/14683 ,  H04N5/2254 ,  H04N5/335

摘要： An image input apparatus has a photoelectric converter element having a flat photosensitive surface and an image formation unit array having a plurality of image formation units arranged in an array. This image formation units individually receive light beams substantially from an identical area and focus the received light beams on different regions of the photosensitive surface of the photoelectric converter element to form images thereon.

摘要翻译： 图像输入装置具有平坦光敏表面的光电转换元件和具有排列成阵列的多个图像形成单元的图像形成单元阵列。 该图像形成单元基本上从相同的区域接收光束，并将接收的光束聚焦在光电转换元件的感光表面的不同区域上，以在其上形成图像。

公开(公告)号：US20090225203A1

公开(公告)日：2009-09-10

申请号：US12395971

申请日：2009-03-02

申请人： Jun Tanida ,  Ryoichi Horisaki ,  Takashi Toyoda ,  Yoshizumi Nakao ,  Yasuo Masaki

发明人： Jun Tanida ,  Ryoichi Horisaki ,  Takashi Toyoda ,  Yoshizumi Nakao ,  Yasuo Masaki

IPC分类号： H04N5/225

CPC分类号： H04N5/2254 ,  G02B3/0056 ,  H04N5/3415

摘要： A compound-eye imaging device comprises an imaging device body having 9 optical lenses and a solid-state imaging element for imaging unit images formed by the optical lenses. Assuming that the combination of each of the optical lenses with a corresponding divided area of the solid-state imaging element to image each of the corresponding unit images is an imaging unit, thereby forming multiple imaging units, the respective imaging units have randomly different optical imaging conditions. For example, the focal lengths of the 9 optical lenses are set to have random values in which the optical lenses are arranged to have random distances between adjacent ones thereof in a direction parallel to the major surface of the solid-state imaging element. This compound-eye imaging device substantially prevents unit images formed by respective imaging units from being the same, making it possible to easily increase the definition of a reconstructed image.

摘要翻译： 复眼成像装置包括具有9个光学透镜的成像装置主体和用于对由光学透镜形成的单元图像进行成像的固态成像元件。 假设每个光学透镜与固态成像元件的对应的分割区域的组合以对相应的单位图像进行成像，则成像单元，从而形成多个成像单元，各个成像单元具有随机地不同的光学成像 条件。 例如，9个光学透镜的焦距被设定为具有随机值，其中光学透镜被布置成在与固态成像元件的主表面平行的方向上具有相邻光学透镜之间的随机距离。 该复眼成像装置基本上防止由各个成像单元形成的单位图像相同，使得可以容易地增加重建图像的清晰度。

公开(公告)号：US07657122B2

公开(公告)日：2010-02-02

申请号：US10596137

申请日：2004-12-01

申请人： Jun Tanida ,  Kouichi Nitta

发明人： Jun Tanida ,  Kouichi Nitta

IPC分类号： G06K9/32 ,  G06K9/36

CPC分类号： G06T3/40 ,  H04N5/2254 ,  H04N5/335

摘要： Using image data on a plurality of reduced object images to calculate a shift amount in regard to the gap of relative positions between the reduced object images by a correlation calculation between the reduced object images. A conversion equation is obtained from the shift amount for geometric projection process from the object image to each of the reduced object images. Initial image data on a single object image is generated using image data on the plurality of reduced object images. The above are used to estimate images of each of the reduced object images. A difference between the estimated image of the reduced object images and the reduced object images is projected in the reverse process of the geometric projection process, updating the image data on the single object image. The processes are repeated until the difference satisfies a predetermined condition and a high resolution object image is outputted.

摘要翻译： 使用多个缩小对象图像上的图像数据，通过缩小对象图像之间的相关计算来计算关于缩小对象图像之间的相对位置的间隙的偏移量。 从对象图像到每个缩小对象图像的几何投影处理的偏移量获得转换方程。 使用多个缩小对象图像上的图像数据来生成单个对象图像上的初始图像数据。 上述用于估计每个缩小的对象图像的图像。 在几何投影处理的相反过程中投影缩小对象图像的估计图像和缩小对象图像之间的差异，更新单个对象图像上的图像数据。 重复这些处理，直到差异满足预定条件并输出高分辨率物体图像。

公开(公告)号：US20090060281A1

公开(公告)日：2009-03-05

申请号：US12261706

申请日：2008-10-30

申请人： Jun TANIDA ,  Takashi Toyoda ,  Yoshizumi Nakao ,  Yasuo Masaki

发明人： Jun TANIDA ,  Takashi Toyoda ,  Yoshizumi Nakao ,  Yasuo Masaki

IPC分类号： G06K9/00

CPC分类号： G06T3/00 ,  H04N5/3415

摘要： An object distance deriving device comprises a compound-eye imaging unit for capturing n unit images and a microprocessor for calculating an object distance of an object from the imaging unit based on the unit images. The microprocessor sets a first temporary distance D1 from discrete temporary distances D1-Dn prepared in advance, and rearranges pixels of each unit image at D1 to create one reconstructed image. The microprocessor reversely projects the pixels of each unit image at D1 to create n reverse projection images. The microprocessor calculates and sums n deviations each between a pixel of the reconstructed image and that of each reverse projection image at each xy coordinate position to calculate an evaluation value for D1. The microprocessor repeats this process for the temporary distances D2-Dn to obtain n evaluation values. The microprocessor determines one of the temporary distances D1-Dn giving a minimum evaluation value as the object distance.

摘要翻译： 物体距离导出装置包括用于捕获n个单位图像的复眼成像单元和用于基于单位图像计算物体与成像单元的物体距离的微处理器。 微处理器预先准备的离散临时距离D1-Dn设置第一临时距离D1，并且在D1处重新排列每个单位图像的像素以创建一个重建图像。 微处理器在D1处反向投影每个单位图像的像素，以产生n个反投影图像。 微处理器在每个xy坐标位置处计算和求和重建图像的像素与每个反向投影图像的像素之间的n个偏差，以计算D1的评估值。 微处理器对该临时距离D2-Dn重复该处理以获得n个评估值。 微处理器确定给出最小评估值的临时距离D1-Dn中的一个作为对象距离。

公开(公告)号：US20080247638A1

公开(公告)日：2008-10-09

申请号：US12055762

申请日：2008-03-26

申请人： Jun Tanida ,  Takashi Toyoda ,  Yoshizumi Nakao ,  Yasuo Masaki

发明人： Jun Tanida ,  Takashi Toyoda ,  Yoshizumi Nakao ,  Yasuo Masaki

IPC分类号： G06T15/00

CPC分类号： G06T7/557

摘要： A three-dimensional object imaging device comprises a compound-eye imaging unit and an image reconstructing unit for reconstructing an image of a three-dimensional object based on multiple unit images captured by the imaging unit. Based on the unit images obtained by the imaging unit, the image reconstructing unit calculates a distance (hereafter “pixel distance”) between the object and the imaging unit for each pixel forming the unit images, and rearranges the unit images pixel-by-pixel on a plane at the pixel distance to create a reconstructed image. Preferably, the image reconstructing unit sums a high-frequency component reconstructed image created from the multiple unit images with a lower noise low-frequency component unit image selected from low-frequency component unit images created from the multiple unit images so as to form a reconstructed image of the three-dimensional object. This makes it possible to obtain a reconstructed image with high definition easily by a simple process.

摘要翻译： 一种三维物体摄像装置，其特征在于，具备：复眼图像单元和图像重构单元，用于基于由所述摄像单元拍摄的多个单位图像来重构三维物体的图像。 基于由成像单元获得的单位图像，图像重建单元针对形成单位图像的每个像素计算对象和成像单元之间的距离（以下称为“像素距离”），并且逐个像素地重新排列单位图像 在像素距离的平面上创建重建图像。 优选地，图像重建单元将从多个单位图像创建的高频分量重构图像与从多个单位图像创建的低频分量单元图像中选择的较低噪声低频分量单元图像相加，以形成重构的 三维物体的图像。 这使得可以通过简单的过程容易地获得具有高清晰度的重建图像。

公开(公告)号：US08237841B2

公开(公告)日：2012-08-07

申请号：US12395971

申请日：2009-03-02

申请人： Jun Tanida ,  Ryoichi Horisaki ,  Takashi Toyoda ,  Yoshizumi Nakao ,  Yasuo Masaki

发明人： Jun Tanida ,  Ryoichi Horisaki ,  Takashi Toyoda ,  Yoshizumi Nakao ,  Yasuo Masaki

IPC分类号： H04N5/225 ,  G02B13/16

CPC分类号： H04N5/2254 ,  G02B3/0056 ,  H04N5/3415

摘要： A compound-eye imaging device comprises an imaging device body having 9 optical lenses and a solid-state imaging element for imaging unit images formed by the optical lenses. Assuming that the combination of each of the optical lenses with a corresponding divided area of the solid-state imaging element to image each of the corresponding unit images is an imaging unit, thereby forming multiple imaging units, the respective imaging units have randomly different optical imaging conditions. For example, the focal lengths of the 9 optical lenses are set to have random values in which the optical lenses are arranged to have random distances between adjacent ones thereof in a direction parallel to the major surface of the solid-state imaging element. This compound-eye imaging device substantially prevents unit images formed by respective imaging units from being the same, making it possible to easily increase the definition of a reconstructed image.

摘要翻译： 复眼成像装置包括具有9个光学透镜的成像装置主体和用于对由光学透镜形成的单元图像进行成像的固态成像元件。 假设每个光学透镜与固态成像元件的对应的分割区域的组合以对相应的单位图像进行成像，则成像单元，从而形成多个成像单元，各个成像单元具有随机地不同的光学成像 条件。 例如，9个光学透镜的焦距被设定为具有随机值，其中光学透镜被布置成在与固态成像元件的主表面平行的方向上具有相邻光学透镜之间的随机距离。 该复眼成像装置基本上防止由各个成像单元形成的单位图像相同，使得可以容易地增加重建图像的清晰度。

公开(公告)号：US08115156B2

公开(公告)日：2012-02-14

申请号：US12551088

申请日：2009-08-31

申请人： Jun Tanida ,  Ryoichi Horisaki ,  Yoshizumi Nakao ,  Takashi Toyoda ,  Yasuo Masaki

发明人： Jun Tanida ,  Ryoichi Horisaki ,  Yoshizumi Nakao ,  Takashi Toyoda ,  Yasuo Masaki

IPC分类号： H01L27/00

CPC分类号： H04N5/2259 ,  H01L27/14625 ,  H01L27/14685 ,  H04N5/2254 ,  H04N5/349

摘要： An imaginary object plane is set in front of an imaging device body (plane setting step). A part of optical conditions of optical lenses are changed as variables, and positions of points (pixel observation points) on the imaginary object plane where lights coming from pixels of a solid-state imaging element and back-projected through the optical lenses are calculated (pixel observation point calculating step). The dispersion in position of the calculated pixel observation points is evaluated (evaluating step). Finally, a set of values of the variables giving maximum evaluated dispersion of the calculated pixel observation points is determined as optimum optical condition of the optical lenses (condition determining step). This reduces the number of pixels which image the same portions of the target object, making it possible to reduce portions of the same image information in multiple unit images, and to stably obtain a reconstructed image having a high definition.

摘要翻译： 将虚拟物平面设置在成像装置主体的前方（平面设置步骤）。 光学透镜的光学条件的一部分作为变量而变化，并且在计算出来自固态成像元件的像素并通过光学透镜反投影的光的虚拟物面上的点（像素观察点）的位置（ 像素观察点计算步骤）。 评估计算出的像素观察点的位置偏差（评价步骤）。 最后，将给出计算出的像素观察点的最大评估色散的变量的值的集合确定为光学透镜的最佳光学条件（条件确定步骤）。 这减少了对目标对象的相同部分进行成像的像素数量，从而可以减少多个单位图像中相同图像信息的部分，并且稳定地获得具有高清晰度的重建图像。