WO2019119452A1 - 图像传感器及其获取图像的方法、智能设备 - Google Patents

图像传感器及其获取图像的方法、智能设备 Download PDF

Info

Publication number
WO2019119452A1
WO2019119452A1 PCT/CN2017/118119 CN2017118119W WO2019119452A1 WO 2019119452 A1 WO2019119452 A1 WO 2019119452A1 CN 2017118119 W CN2017118119 W CN 2017118119W WO 2019119452 A1 WO2019119452 A1 WO 2019119452A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
image sensor
layer
filter
control command
Prior art date
Application number
PCT/CN2017/118119
Other languages
English (en)
French (fr)
Inventor
阳光
Original Assignee
深圳配天智能技术研究院有限公司
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 深圳配天智能技术研究院有限公司 filed Critical 深圳配天智能技术研究院有限公司
Priority to CN201780035379.1A priority Critical patent/CN109314746B/zh
Priority to PCT/CN2017/118119 priority patent/WO2019119452A1/zh
Publication of WO2019119452A1 publication Critical patent/WO2019119452A1/zh

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/60—Control of cameras or camera modules

Definitions

  • the present invention relates to the field of vision systems, and in particular to an image sensor and a method thereof, and a smart device.
  • image sensors are the most critical components of smart devices, such as digital cameras, camera image devices, for converting optical images that are focused by a lens into digital images.
  • the image sensor is capable of converting the brightness of visible light that is incident on the sensor into an electrical signal and digitizing it in subsequent processing to produce a grayscale image.
  • a color filter is usually disposed on the image sensor, so that the image sensor can digitize light of different colors in visible light to generate a color image.
  • the color filters are distributed according to the common RGB arrangement, and there may be problems of poor local or edge color effects, such as low image sharpness and low resolution.
  • the technical problem to be solved by the present invention is to provide an image sensor and a method and an intelligent device for acquiring an image, which improve the sharpness, color richness and resolution of the acquired image, and improve the user experience.
  • the first technical solution adopted by the present invention is to provide an image sensor, the image sensor including a sensor layer and a filter layer, the sensor layer including at least one image acquisition area, the filter
  • the layer includes at least one filter region;
  • the image sensor further includes a processor, the filter layer and the sensor layer are respectively coupled to the processor;
  • the sensor layer is configured to acquire a reference image; Generating a control command according to the reference image, and transmitting the control command to the filter layer;
  • the filter layer is configured to adjust a light transmittance of the filter region according to the control command, to obtain a
  • the photosensitive distribution matched by the control command is controlled to cause the image sensor to acquire an image in accordance with the photosensitive distribution.
  • the second technical solution adopted by the present invention is to provide a smart device, which comprises the image sensor according to any of the inventions.
  • the third technical solution adopted by the present invention is to provide a method for acquiring an image, the method for acquiring an image includes: an image sensor acquiring a reference image through a sensor layer thereof; The reference image generates a control command; and adjusts the light transmittance of the filter region according to the control command to obtain a light-sensing distribution matched with the control command, so that the image sensor collects an image according to the photosensitive distribution.
  • the invention has the beneficial effects that the filter layer of the image sensor of the invention can control the light transmittance of the filter region according to the control instruction of the processor, so as to adjust the photosensitive distribution of the filter layer, and can selectively set the filter according to the actual situation.
  • the light-sensing distribution of the light layer enables the image sensor to acquire images according to the photosensitive distribution, which improves the sharpness, color richness and resolution of the acquired image.
  • FIG. 1 is a schematic structural view of an embodiment of an image sensor of the present invention
  • FIG. 2 is a schematic flow chart of an embodiment of a method for acquiring an image according to the present invention.
  • the present invention provides an image sensor and a method and an intelligent device for acquiring an image.
  • an image sensor and a method and an intelligent device for acquiring an image.
  • the present invention will be further described in detail below, and the specific implementation regulations described herein should be understood. It is intended to explain the invention only and is not intended to limit the invention.
  • the image sensor of the present embodiment includes a sensor layer and a filter layer, wherein the sensor layer includes at least one image acquisition region, and the filter layer includes at least one filter region.
  • the image sensor also includes a processor, and the filter layer and the sensor layer are coupled to the processor, respectively.
  • the sensor layer is configured to acquire a reference image
  • the processor is configured to generate a control instruction according to the reference image, and send the control instruction to the filter layer
  • the filter layer is configured to adjust the light transmittance of the filter region according to the control instruction, and obtain The photosensitive distribution matched by the command is controlled to cause the image sensor to acquire an image in accordance with the photosensitive distribution.
  • FIG. 1 is a schematic structural view of an image sensor according to an embodiment of the present invention.
  • the direction indicated by the arrow in FIG. 1 is the direction in which the light is incident.
  • the image sensor 10 of the present embodiment includes a filter layer 101 and a sensor layer 102.
  • the image sensor 10 further includes a processor 103 coupled to the filter layer 101 and the sensor layer 102, respectively.
  • the image sensor 10 also includes an isolation layer between the sensor layer 102 and the filter layer 101.
  • the material of the isolation layer is a dielectric material such as silicon oxide, silicon nitride or silicon oxynitride for isolating the filter layer 101 and the sensor layer 102.
  • the sensor layer 102 includes at least one image acquisition area, and the number of image acquisition areas is not specifically limited, and may be designed according to actual conditions, for example, including 2 or 3 or 6 image acquisition areas.
  • the image acquisition area is used to convert the received optical signal into an electrical signal.
  • the filter layer 101 includes at least one filter region, and the number of filter regions is not specifically limited, and may be designed according to actual conditions, for example, including 2 or 3 or 6 filter regions.
  • the filter area is used to select a specific wavelength of light to pass through, so that the corresponding image acquisition area can acquire the corresponding pixel color.
  • the filter region is formed by a photonic crystal that is periodically arranged by media of different refractive indices, and is capable of controlling its characteristics according to a voltage, thereby selecting light of a certain band to pass.
  • the filter region can also be formed by a grating device that is an electrically controlled diffraction grating device that is capable of controlling its characteristics according to a voltage to select a certain band of light to pass through.
  • the number of filter regions matches the number of image acquisition regions, and the distribution of the filter regions matches the distribution of the image acquisition regions.
  • the number of the filtering regions is equal to the number of the image capturing regions, and the position of each image capturing region is The positions of each of the filter regions are in one-to-one correspondence, so that the photosensitive distribution of the filter layer 101 can be precisely adjusted.
  • the filter area and the image collection area may be designed according to actual conditions, and are not specifically limited herein.
  • the sensor layer 102 is used to acquire a reference image, and the image acquisition area in the sensor layer 102 converts the received optical signal into an electrical signal, and digitally processes the electrical signal according to the intensity of the light to obtain a reference image.
  • the filter layer 101 is preset with a specific photosensitive distribution, and the sensor layer 102 acquires an image as a reference image according to the preset photosensitive distribution.
  • the filter layer 101 is used to perform a filtering operation on the light irradiated on the image sensor 10. Specifically, the filter regions in the filter layer 101 selectively transmit light of a corresponding wavelength range while blocking light transmission of other wavelength ranges to obtain color information.
  • the processor 103 is configured to receive a reference image of the sensor layer 102, generate a control instruction according to the reference image, and send the control command to the filter layer 101, so that the filter layer 10 adjusts the filter region according to the control instruction.
  • the light transmittance is such that a photosensitive distribution matching the control command is obtained to cause the image sensor 10 to acquire an image in accordance with the adjusted photosensitive distribution.
  • the processor 103 analyzes the reference image using a Bayesian algorithm to generate a control command and transmits the control command to the filter layer 101.
  • the Bayesian algorithm uses the inverse push method to determine the probability of occurrence of an event, thereby determining the optimal photosensitive distribution of the filter layer 10.
  • the Bayesian algorithm is based on inference to determine the probability of an event occurring.
  • the reasoning is divided into two processes.
  • the first step is to establish a model for the observed data.
  • the second step is to use this model to estimate the probability of an unknown phenomenon occurring.
  • the image sensor 10 builds a model from the reference image, analyzes the model, estimates the possibility of multiple photosensitive distributions, and verifies which photosensitive distribution can best correspond to the target object corresponding to the reference image. The actual color to determine the optimum sensitometric distribution.
  • the filter distribution of the filter layer of the image sensor such as the RGB distribution
  • the filter distribution of the filter layer of the image sensor is preset in the factory, so it is impossible to reset the filter distribution of the filter layer according to the actual situation, resulting in the edge position of the acquired image or other
  • the sharpness and resolution of the image are not high, and the color of the image is not rich enough.
  • the photosensitive distribution of the filter layer 101 of the present embodiment may be changed according to actual conditions.
  • the filter layer 101 is pre-set with a first photosensitive distribution
  • the sensor layer 102 acquires a reference image through the first photosensitive distribution.
  • the processor 103 analyzes the reference image, generates a control command according to the result of the processing, and sends the control command to the filter layer 101, so that the filter layer 101 is adjusted according to the control command.
  • the light transmission of the filter region gives a photosensitive distribution that matches the control command.
  • the processor 103 processes the reference image by using a Bayesian algorithm, generates a voltage control instruction according to the result of the processing, and sends the voltage control instruction to the filter layer 101 to filter
  • the optical layer 101 adjusts the wavelength range of the light that can be transmitted by the filter region according to the voltage control command, and obtains a photosensitive distribution that matches the voltage control command.
  • the voltage control command includes a voltage corresponding to the filter region, and the wavelength range of the light that can be transmitted by the filter region can be adjusted according to the voltage.
  • the color of the visible light that can be transmitted by the filter region in the filter layer 101 can be red or blue or green, that is, the filter region can transmit visible light in a wavelength range corresponding to red or blue or green.
  • the red visible light corresponds to a wavelength range of 620 nm to 750 nm
  • the blue visible light corresponds to a wavelength range of 450 nm to 475 nm
  • the green visible light corresponds to a wavelength range of 495 nm to 570 nm.
  • the processor 103 can inversely obtain the position of the reference image that needs to be adjusted and the color corresponding to the position, thereby determining the filter area to be adjusted and the filter.
  • the light region corresponds to the wavelength of the light to be transmitted, and the voltage applied to the filter region can be determined according to the wavelength, thereby generating a control command to adjust the photosensitive distribution of the filter layer 101, so that the image sensor 10 is collected according to the adjusted photosensitive distribution. image.
  • the edge position of the image is prone to loss of color information and blurring of the image.
  • the image sensor 10 can adjust only the photosensitive distribution of the filter regions corresponding to the edge positions of the sensor layer 101, thereby improving the sharpness of the image edge positions.
  • the image sensor 10 can adjust the photosensitive distribution of the sensor layer 101 at least once.
  • the number of specific adjustments can be designed according to the actual situation, for example, one or three times, which is not specifically limited, and the image needs to be satisfied.
  • the performance indicators can be used.
  • the filter layer of the image sensor of the present embodiment can control the light transmittance of the filter region according to the control instruction of the processor to adjust the photosensitive distribution of the filter layer, and can selectively set the filter according to actual conditions.
  • the light-sensing distribution of the light layer enables the image sensor to acquire images according to the photosensitive distribution, which improves the sharpness, color richness and resolution of the acquired image.
  • FIG. 2 is a schematic flow chart of an embodiment of a method for acquiring an image according to the present invention.
  • the method of acquiring an image of the present embodiment is applied to the image sensor of any of the above embodiments.
  • the method for obtaining an image includes:
  • the image sensor acquires a reference image through its sensor layer.
  • the image sensor of the present embodiment includes a sensor layer and a filter layer.
  • the image sensor also includes a processor coupled to the filter layer and the sensor layer, respectively.
  • the sensor layer includes at least one image acquisition area, and the number of image collection areas is not specifically limited, and may be designed according to actual conditions, for example, including 2 or 3 or 6 image acquisition areas.
  • the image acquisition area is used to convert the received optical signal into an electrical signal.
  • the filter layer includes at least one filter region, and the number of filter regions is not specifically limited, and may be designed according to actual conditions, for example, including 2 or 3 or 6 filter regions.
  • the filter area is used to select a specific wavelength of light to pass through, so that the corresponding image acquisition area can acquire the corresponding pixel color.
  • the image sensor acquires a reference image through its sensor layer.
  • the image acquisition area in the sensor layer converts the received optical signal into an electrical signal, and digitally processes the electrical signal according to the intensity of the light to obtain a reference image.
  • the image sensor generates a control instruction according to the reference image.
  • the processor of the image sensor receives the reference image of the sensor layer, generates a control command according to the reference image, and sends the control command to the filter layer, so that the filter layer adjusts the filter region according to the control command.
  • the light transmittance is such that a photosensitive distribution matching the control command is obtained to cause the image sensor to acquire an image according to the adjusted photosensitive distribution.
  • the processor processes the reference image by using a Bayesian algorithm, generates a voltage control instruction according to the processing result, and sends the voltage control instruction to the filter layer to enable the filter layer
  • the wavelength range of the light that can be transmitted by the filter region is adjusted according to the voltage control command, and a light-sensing distribution matching the voltage control command is obtained.
  • the voltage control command includes a voltage corresponding to the filter region, and the wavelength range of the light that can be transmitted by the filter region can be adjusted according to the voltage.
  • the Bayesian algorithm uses the inverse method to determine the probability of occurrence of an event, thereby determining the optimal photosensitive distribution of the filter layer.
  • the Bayesian algorithm is based on inference to determine the probability of an event occurring.
  • the reasoning is divided into two processes.
  • the first step is to establish a model for the observed data.
  • the second step is to use this model to estimate the probability of an unknown phenomenon occurring.
  • the image sensor establishes a model through the reference image, analyzes the model, estimates the possibility of multiple photosensitive distributions, and verifies which photosensitive distribution can best correspond to the actual target object corresponding to the reference image. Color to determine the best sensitization distribution.
  • the color of the visible light that can be transmitted through the filter region in the filter layer can be red or blue or green, that is, the filter region can transmit visible light in a wavelength range corresponding to red or blue or green.
  • the red visible light corresponds to a wavelength range of 620 nm to 750 nm
  • the blue visible light corresponds to a wavelength range of 450 nm to 475 nm
  • the green visible light corresponds to a wavelength range of 495 nm to 570 nm.
  • the processor processes the reference image by using the Bayesian algorithm, the position of the reference image to be adjusted and the color corresponding to the position are reversibly obtained, thereby determining the filter region to be adjusted and the filter.
  • the region corresponds to the wavelength of the light that needs to be transmitted, from which the voltage applied to the filter region can be determined to generate a control command.
  • the image sensor adjusts the light transmittance of the filter region according to the control command to obtain a light-sensing distribution that matches the control command, so that the image sensor collects the image according to the light-sensing distribution.
  • the filter region is formed by a photonic crystal that is periodically arranged by media of different refractive indices, and is capable of controlling its characteristics according to a voltage, thereby selecting light of a certain band to pass.
  • the filter region can also be formed by a grating device that is an electrically controlled diffraction grating device that is capable of controlling its characteristics according to a voltage to select a certain band of light to pass through.
  • control command is a voltage control command
  • the image sensor adjusts a wavelength range of light that can be transmitted by the filter region according to the voltage control command to obtain a photosensitive distribution matched with the voltage control command, so that The image sensor acquires an image in accordance with the photosensitive distribution.
  • the edge position of the image is prone to loss of color information and blurring of the image.
  • the image sensor can only adjust the light-sensing distribution of the filter region corresponding to the edge position of the sensor layer, thereby improving the sharpness of the image edge position.
  • the image sensor can adjust the photosensitive distribution of the sensor layer at least once, and the specific number of adjustments can be designed according to actual conditions, for example, one or three adjustments, which are not specifically limited, and need to meet image performance indicators. The needs can be.
  • the filter layer of the image sensor of the present embodiment can control the light transmittance of the filter region according to the control instruction of the processor to adjust the photosensitive distribution of the filter layer, and can selectively set the filter according to the actual situation.
  • the photosensitive distribution of the layer is such that the image sensor collects images according to the photosensitive distribution, which improves the sharpness, color richness and resolution of the acquired image.
  • a smart device comprising the image sensor of any of the above embodiments.
  • smart devices include cameras, industrial robots, video cameras or smart phones.
  • the filter layer of the image sensor of the smart device of the present embodiment can control the light transmittance of the filter region according to the control instruction of the processor to adjust the photosensitive distribution of the filter layer, and can be selected according to actual conditions.
  • the photosensitive distribution of the filter layer is set to enable the image sensor to collect images according to the photosensitive distribution, thereby improving the sharpness, color richness and resolution of the acquired image.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Color Television Image Signal Generators (AREA)
  • Transforming Light Signals Into Electric Signals (AREA)
  • Solid State Image Pick-Up Elements (AREA)

Abstract

本发明公开了提供一种图像传感器及其获取图像的方法、智能设备，该图像传感器包括传感器层以及滤光层，传感器层包括至少一个图像采集区域，滤光层包括至少一个滤光区域；图像传感器还包括处理器，滤光层以及传感器层分别与处理器耦接；传感器层用于获取基准图像；处理器用于根据基准图像生成控制指令，并将控制指令发送给滤光层；滤光层用于根据控制指令调整滤光区域的透光性，得到与控制指令匹配的感光分布，以使图像传感器按照感光分布采集图像。该图像传感器提升了获取到的图像的锐度、颜色丰富度及分辨率。

Description

图像传感器及其获取图像的方法、智能设备
【技术领域】
本发明涉及视觉系统领域，特别是涉及一种图像传感器及其获取图像的方法、智能设备。
【背景技术】
在视觉系统领域，图像传感器是智能设备，如数码相机、摄像机的图像设备中最关键的部件，用于将镜头聚焦的光学图像转换为数字图像。图像传感器能够将照射到传感器上的可见光的亮度转换为电信号，并在后续的处理中数字化，产生灰度图像。
为了获取自然界的颜色信息产生彩色图像，通常在图像传感器上设置滤色片，使得图像传感器能够对可见光中不同颜色的光进行数字化，产生彩色图像。目前，滤色片是按照普通的RGB排列方式分布的，会出现局部或边缘颜色效果差的问题，比如图像锐度低以及分辨率低的问题。
【发明内容】
本发明主要解决的技术问题是提供一种图像传感器及其获取图像的方法、智能设备，提升了获取到的图像的锐度、颜色丰富度及分辨率，提高了用户体验。
为解决上述技术问题，本发明采用的第一个技术方案是：提供一种图像传感器，所述图像传感器包括传感器层以及滤光层，所述传感器层包括至少一个图像采集区域，所述滤光层包括至少一个滤光区域；所述图像传感器还包括处理器，所述滤光层以及所述传感器层分别与所述处理器耦接；所述传感器层用于获取基准图像；所述处理器用于根据所述基准图像生成控制指令，并将所述控制指令发送给所述滤光层；所述滤光层用于根据所述控制指令调整所述滤光区域的透光性，得到与所述控制指令匹配的感光分布，以使所述图像传感器按照所述感光分布采集图像。
为解决上述技术问题，本发明采用的第二个技术方案是：提供一种智能设备，所述智能设备包括本发明任一所述的图像传感器。
为解决上述技术问题，本发明采用的第三个技术方案是：提供一种获取图像的方法，所述获取图像的方法包括：图像传感器通过其传感器层获取基准图像；所述图像传感器根据所述基准图像生成控制指令；并根据所述控制指令调整其滤光区域的透光性，得到与所述控制指令匹配的感光分布，以使所述图像传感器按照所述感光分布采集图像。
本发明的有益效果是：本发明的图像传感器的滤光层可根据处理器的控制指令控制其滤光区域的透光性，以调整滤光层的感光分布，能够根据实际情况选择性设置滤光层的感光分布，以使图像传感器按照感光分布采集图像，提升了获取到的图像的锐度、颜色丰富度及分辨率。
【附图说明】
图1是本发明图像传感器一实施方式的结构示意图；
图2是本发明获取图像的方法一实施方式的流程示意图。
【具体实施方式】
本发明提供一种图像传感器及其获取图像的方法、智能设备，为使本发明的目的、技术方案和技术效果更加明确、清楚，以下对本发明进一步详细说明，应当理解此处所描述的具体实施条例仅用于解释本发明，并不用于限定本发明。
本实施方式的图像传感器包括传感器层以及滤光层，其中，传感器层包括至少一个图像采集区域，滤光层包括至少一个滤光区域。图像传感器还包括处理器，滤光层以及传感器层分别与处理器耦接。
具体地，传感器层用于获取基准图像，处理器用于根据基准图像生成控制指令，并将控制指令发送给滤光层，滤光层用于根据控制指令调整滤光区域的透光性，得到与控制指令匹配的感光分布，以使图像传感器按照感光分布采集图像。
为了清楚的说明上述实施方式的图像传感器的结构，请参阅图1，图1是本发明图像传感器一实施方式的结构示意图。其中，图1中箭头所指的方向为光线入射的方向。
本实施方式的图像传感器10包括滤光层101以及传感器层102，图像传感器10还包括处理器103，处理器103分别与滤光层101和传感器层102耦接。图像传感器10还包括隔离层，隔离层位于传感器层102和滤光层101之间。其中，隔离层的材料为介质材料，例如氧化硅、氮化硅或氧氮化硅，用于隔离滤光层101和传感器层102。
另外，传感器层102包括至少一个图像采集区域，图像采集区域的数目不做具体限定，可根据实际情况设计，例如，包括2个或3个或6个图像采集区域。其中，图像采集区域用于将接收到的光信号转换为电信号。
滤光层101包括至少一个滤光区域，滤光区域的数目不做具体限定，可根据实际情况设计，例如，包括2个或3个或6个滤光区域。其中，滤光区域用于选择特定的波长光线通过，以使对应的图像采集区域能够获取到相应的像素颜色。
在其中的一个实施方式中，滤光区域由光子晶体形成，该光子晶体由不同折射率的介质周期性排列而成，能够根据电压控制其特性，从而选择某个波段的光通过。在另一个实施方式中，滤光区域也可以由光栅器件形成，该光栅器件为电控衍射光栅器件，能够根据电压控制其特性，从而选择某个波段的光通过。
可选地，滤光区域的数量与图像采集区域的数量相匹配，滤光区域的分布与图像采集区域的分布相匹配。在其中的一个实施方式中，为了使采集到的目标物体的颜色信息更真实的还原实际目标物体的颜色信息，滤光区域的数量与图像采集区域的数量相等，每一个图像采集区域的位置与每一个滤光区域的位置一一对应，从而可以精确调整滤光层101的感光分布。在其他的实施方式中，滤光区域以及图像采集区域可根据实际的情况设计，在此不做具体限定。
具体地，传感器层102用于获取基准图像，传感器层102中的图像采集区域将接收到的光信号转换为电信号，根据光的强度对该电信号进行数字处理，得到基准图像。其中，滤光层101会预设有一特定的感光分布，传感器层102根据该预设的感光分布而获取图像为基准图像。
滤光层101用于对照射在图像传感器10上的光进行滤光操作。具体地，滤光层101中的滤光区域选择性透射对应波长范围的光，而阻隔其他波长范围的光透射，以获取颜色信息。
进一步地，处理器103用于接收传感器层102的基准图像，根据该基准图像生成控制指令，并将该控制指令发送给滤光层101，以使滤光层10根据该控制指令调整滤光区域的透光性，得到与该控制指令匹配的感光分布，以使图像传感器10按照调整后的感光分布采集图像。
在其中的一个实施方式中，处理器103采用贝叶斯算法对基准图像进行分析，以生成控制指令，并将控制指令发送给滤光层101。其中，贝叶斯算法是采用逆推的方法确定某个事件出现的概率，由此确定滤光层10的最佳感光分布。
下面说明一下贝叶斯算法的具体原理以及分析过程。贝叶斯算法是基于推理确定某个事件出现的概率的，其中，推理分为两个过程，第一步是对观测数据建立一个模型。第二步则是使用这个模型来推测未知现象发生的概率。在本实施方式中，图像传感器10通过基准图像建立模型，并对该模型进行分析，推测多种感光分布的可能性，并验证哪种感光分布可以最佳的对应基准图像所对应的目标物体的实际颜色，从而确定最佳的感光分布。
目前，图像传感器的滤光层的滤光分布，如RGB分布在出厂时已预先设置了，因此不能够根据实际情况重新设置滤光层的滤光分布，导致获取到的图像的边缘位置或其他位置的图像锐度以及分辨率均不高，图像色彩也不够丰富。
为了解决上述问题，本实施方式的滤光层101的感光分布可以根据实际情况改变。在其中的一个实施方式中，滤光层101会预先设置第一感光分布，传感器层102通过该第一感光分布获取基准图像。处理器103在接收到基准图像后会对该基准图像进行分析处理，并根据处理的结果生成控制指令，并将该控制指令发送至滤光层101，以使滤光层101根据该控制指令调整滤光区域的透光性，得到与控制指令匹配的感光分布。
具体地，处理器103在接收到基准图像后，采用贝叶斯算法对基准图像进行处理，并根据处理的结果生成电压控制指令，并将该电压控制指令发送至滤光层101，以使滤光层101根据该电压控制指令调整滤光区域所能透射的光的波长范围，得到与该电压控制指令匹配的感光分布。其中，该电压控制指令包括滤光区域对应的电压，根据该电压可对应调整滤光区域所能透射的光的波长范围。
由于人类视力所能感知的所有颜色均可以通过红（R）、绿（G）、蓝（B）三个颜色通道的变化以及它们的相互叠加得到。所以，可设置滤光层101中的滤光区域能够透射的可见光的颜色为红色或蓝色或绿色，即滤光区域能够透射红色或蓝色或绿色所对应的波长范围的可见光。 其中，红色可见光对应的波长范围为620nm~750nm，蓝色可见光对应的波长范围为450nm~475nm，绿色可见光对应的波长范围为495nm~570nm。
在本实施方式中，处理器103采用贝叶斯算法对基准图像进行处理后，可逆推得到基准图像中需要调整颜色的位置以及该位置对应的颜色，从而确定需要调整的滤光区域以及该滤光区域对应需要透射的光线的波长，根据该波长可以确定施加到滤光区域的电压，从而生成控制指令，以调整滤光层101的感光分布，以使图像传感器10按照调整后的感光分布采集图像。在一个实际的应用场景中，图像的边缘位置容易丢失颜色信息而导致图像模糊。为了节省硬件资源，并提高获取图像的进度，图像传感器10可以仅调整传感器层101的边缘位置对应的滤光区域的感光分布，即可提高图像边缘位置的锐度。
在此，需要说明的是，图像传感器10可对传感器层101的感光分布进行至少一次调整，具体调整的次数可根据实际情况设计，例如调整一次或三次，在此不做具体限定，需满足图像性能指标的需求即可。
区别于现有技术，本实施方式的图像传感器的滤光层可根据处理器的控制指令控制其滤光区域的透光性，以调整滤光层的感光分布，能够根据实际情况选择性设置滤光层的感光分布，以使图像传感器按照感光分布采集图像，提升了获取到的图像的锐度、颜色丰富度及分辨率。
参阅图2，图2是本发明获取图像的方法一实施方式的流程示意图。本实施方式的获取图像的方法适用于上述任一实施方式的图像传感器。该获取图像的方法包括：
201：图像传感器通过其传感器层获取基准图像。
本实施方式的图像传感器包括传感器层以及滤光层。图像传感器还包括处理器，处理器分别与滤光层和传感器层耦接。
其中，传感器层包括至少一个图像采集区域，图像采集区域的数目不做具体限定，可根据实际情况设计，例如，包括2个或3个或6个图像采集区域。其中，图像采集区域用于将接收到的光信号转换为电信号。
滤光层包括至少一个滤光区域，滤光区域的数目不做具体限定，可根据实际情况设计，例如，包括2个或3个或6个滤光区域。其中，滤光区域用于选择特定的波长光线通过，以使对应的图像采集区域能够获取到相应的像素颜色。
在本实施方式中，图像传感器通过其传感器层获取基准图像。
具体地，传感器层中的图像采集区域将接收到的光信号转换为电信号，根据光的强度对该电信号进行数字处理，得到基准图像。
202：图像传感器根据基准图像生成控制指令。
在本实施方式中，图像传感器的处理器接收传感器层的基准图像，根据该基准图像生成控制指令，并将该控制指令发送给滤光层，以使滤光层根据该控制指令调整滤光区域的透光性，得到与该控制指令匹配的感光分布，以使图像传感器按照调整后的感光分布采集图像。
具体地，处理器在接收到基准图像后，采用贝叶斯算法对基准图像进行处理，并根据处理的结果生成电压控制指令，并将该电压控制指令发送至滤光层，以使滤光层根据该电压控制指令调整滤光区域所能透射的光的波长范围，得到与该电压控制指令匹配的感光分布。其中，该电压控制指令包括滤光区域对应的电压，根据该电压可对应调整滤光区域所能透射的光的波长范围。
其中，贝叶斯算法是采用逆推的方法确定某个事件出现的概率，由此确定滤光层的最佳感光分布。
下面说明一下贝叶斯算法的具体原理以及分析过程。
贝叶斯算法是基于推理确定某个事件出现的概率的，其中，推理分为两个过程，第一步是对观测数据建立一个模型。第二步则是使用这个模型来推测未知现象发生的概率。在本实施方式中，图像传感器通过基准图像建立模型，并对该模型进行分析，推测多种感光分布的可能性，并验证哪种感光分布可以最佳的对应基准图像所对应的目标物体的实际颜色，从而确定最佳的感光分布。
由于人类视力所能感知的所有颜色均可以通过红（R）、绿（G）、蓝（B）三个颜色通道的变化以及它们的相互叠加得到。所以，可设置滤光层中的滤光区域能够透射的可见光的颜色为红色或蓝色或绿色，即滤光区域能够透射红色或蓝色或绿色所对应的波长范围的可见光。 其中，红色可见光对应的波长范围为620nm~750nm，蓝色可见光对应的波长范围为450nm~475nm，绿色可见光对应的波长范围为495nm~570nm。
在本实施方式中，处理器采用贝叶斯算法对基准图像进行处理后，可逆推得到基准图像中需要调整颜色的位置以及该位置对应的颜色，从而确定需要调整的滤光区域以及该滤光区域对应需要透射的光线的波长，根据该波长可以确定施加到滤光区域的电压，从而生成控制指令。
203：并根据控制指令调整其滤光区域的透光性，得到与控制指令匹配的感光分布，以使图像传感器按照感光分布采集图像。
在本实施方式中，图像传感器根据控制指令调整其滤光区域的透光性，得到与控制指令匹配的感光分布，以使图像传感器按照感光分布采集图像。
在其中的一个实施方式中，滤光区域由光子晶体形成，该光子晶体由不同折射率的介质周期性排列而成，能够根据电压控制其特性，从而选择某个波段的光通过。在另一个实施方式中，滤光区域也可以由光栅器件形成，该光栅器件为电控衍射光栅器件，能够根据电压控制其特性，从而选择某个波段的光通过。
在其中的一个实施方式中，控制指令为电压控制指令，图像传感器根据电压控制指令调整所述滤光区域所能透射的光的波长范围，得到与所述电压控制指令匹配的感光分布，以使图像传感器按照所述感光分布采集图像。
在一个实际的应用场景中，图像的边缘位置容易丢失颜色信息而导致图像模糊。为了节省硬件资源，并提高获取图像的进度，图像传感器可以仅调整传感器层的边缘位置对应的滤光区域的感光分布，即可提高图像边缘位置的锐度。
在此，需要说明的是，图像传感器可对传感器层的感光分布进行至少一次调整，具体调整的次数可根据实际情况设计，例如调整一次或三次，在此不做具体限定，需满足图像性能指标的需求即可。
区别于现有技术本实施方式的图像传感器的滤光层可根据处理器的控制指令控制其滤光区域的透光性，以调整滤光层的感光分布，能够根据实际情况选择性设置滤光层的感光分布，以使图像传感器按照感光分布采集图像，提升了获取到的图像的锐度、颜色丰富度及分辨率。
进一步地，在另一个实施方式提供了一种智能设备，该智能设备包括上述任一实施方式的图像传感器。
其中，智能设备包括照相机、工业机器人、摄像机或者智能手机。
关于图像传感器的结构以及其获取图像的方法结合上述图1和图2以及相关的文字，已详尽描述，在此不再赘谈。
区别于现有技术，本实施方式的智能设备的图像传感器的滤光层可根据处理器的控制指令控制其滤光区域的透光性，以调整滤光层的感光分布，能够根据实际情况选择性设置滤光层的感光分布，以使图像传感器按照感光分布采集图像，提升了获取到的图像的锐度、颜色丰富度及分辨率。以上所述仅为本发明的实施方式，并非因此限制本发明的专利保护范围，凡是利用本发明说明书及附图内容所作的等效结构或等效流程变换，或直接或间接运用在其他相关的技术领域，均同理包括在本发明的专利保护范围内。

Claims (14)

  1. 一种图像传感器，其特征在于，所述图像传感器包括传感器层以及滤光层，所述传感器层包括至少一个图像采集区域，所述滤光层包括至少一个滤光区域；
    所述图像传感器还包括处理器，所述滤光层以及所述传感器层分别与所述处理器耦接；
    所述传感器层用于获取基准图像；
    所述处理器用于根据所述基准图像生成控制指令，并将所述控制指令发送给所述滤光层；
    所述滤光层用于根据所述控制指令调整所述滤光区域的透光性，得到与所述控制指令匹配的感光分布，以使所述图像传感器按照所述感光分布采集图像。
  2. 根据权利要求1所述的图像传感器，其特征在于，所述处理器具体用于采用贝叶斯算法对所述基准图像进行分析，以生成控制指令，并将所述控制指令发送给所述滤光层。
  3. 根据权利要求1所述的图像传感器，其特征在于，所述滤光区域由光子晶体形成，其中，所述光子晶体由不同折射率的介质周期性排列而成，能够根据电压选择某个波段的光通过。
  4. 根据权利要求1所述的图像传感器，其特征在于，所述滤光区域由光栅器件形成，其中，所述光栅器件为电控衍射光栅器件，能够根据电压选择某个波段的光通过。
  5. 根据权利要求3或4所述的图像传感器，其特征在于，所述控制指令为电压控制指令，所述滤光层用于根据所述控制指令调整所述滤光区域的透光性，得到与所述控制指令匹配的感光分布具体为：
    所述滤光层用于根据所述电压控制指令调整所述滤光区域所能透射的光的波长范围，得到与所述电压控制指令匹配的感光分布。
  6. 根据权利要求5所述的图像传感器，其特征在于，所述波长范围为620nm~750nm或495nm~570nm或450nm~475nm中的一种。
  7. 根据权利要求1所述的图像传感器，其特征在于，所述图像传感器还包括隔离层，所述隔离层位于所述传感器层和所述滤光层之间。
  8. 一种智能设备，其特征在于，所述智能设备包括如权利要求1-7任一所述的图像传感器。
  9. 一种获取图像的方法，其特征在于，所述获取图像的方法包括：
    图像传感器通过其传感器层获取基准图像；
    所述图像传感器根据所述基准图像生成控制指令；
    并根据所述控制指令调整其滤光区域的透光性，得到与所述控制指令匹配的感光分布，以使所述图像传感器按照所述感光分布采集图像。
  10. 根据权利要求9所述的获取图像的方法，其特征在于，所述图像传感器根据所述基准图像生成控制指令的步骤具体为：
    所述图像传感器采用贝叶斯算法对所述基准图像进行分析，以生成控制指令。
  11. 根据权利要求9所述的获取图像的方法，其特征在于，所述滤光区域由光子晶体形成，其中，所述光子晶体由不同折射率的介质周期性排列而成，能够根据电压选择某个波段的光通过。
  12. 根据权利要求9所述的获取图像的方法，其特征在于，所述滤光区域由光栅器件形成，其中，所述光栅器件为电控衍射光栅器件，能够根据电压选择某个波段的光通过。
  13. 根据权利要求11或12所述的获取图像的方法，其特征在于，所述控制指令为电压控制指令，所述并根据所述控制指令调整其滤光区域的透光性，得到与所述控制指令匹配的感光分布，以使所述图像传感器按照所述感光分布采集图像的步骤具体包括：
    所述图像传感器根据所述电压控制指令调整其滤光区域所能透射的光的波长范围，得到与所述电压控制指令匹配的感光分布，以使所述图像传感器按照所述感光分布采集图像。
  14. 根据权利要求13所述的获取图像的方法，其特征在于，所述波长范围为620nm~750nm或495nm~570nm或450nm~475nm中的一种。
PCT/CN2017/118119 2017-12-22 2017-12-22 图像传感器及其获取图像的方法、智能设备 WO2019119452A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN201780035379.1A CN109314746B (zh) 2017-12-22 2017-12-22 图像传感器及其获取图像的方法、智能设备
PCT/CN2017/118119 WO2019119452A1 (zh) 2017-12-22 2017-12-22 图像传感器及其获取图像的方法、智能设备

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2017/118119 WO2019119452A1 (zh) 2017-12-22 2017-12-22 图像传感器及其获取图像的方法、智能设备

Publications (1)

Publication Number Publication Date
WO2019119452A1 true WO2019119452A1 (zh) 2019-06-27

Family

ID=65225726

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/118119 WO2019119452A1 (zh) 2017-12-22 2017-12-22 图像传感器及其获取图像的方法、智能设备

Country Status (2)

Country Link
CN (1) CN109314746B (zh)
WO (1) WO2019119452A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116913937A (zh) * 2023-08-25 2023-10-20 浙桂(杭州)半导体科技有限责任公司 图像传感器件及其光学调控方法

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20120127334A1 (en) * 2010-11-18 2012-05-24 Canon Kabushiki Kaisha Adaptive spatial sampling using an imaging assembly having a tunable spectral response
CN103491319A (zh) * 2012-06-11 2014-01-01 索尼公司 成像装置
US8836808B2 (en) * 2011-04-19 2014-09-16 Canon Kabushiki Kaisha Adaptive color imaging by using an imaging assembly with tunable spectral sensitivities
US20140293091A1 (en) * 2012-05-21 2014-10-02 Digimarc Corporation Sensor-synchronized spectrally-structured-light imaging
US8854515B2 (en) * 2011-09-09 2014-10-07 Canon Kabushiki Kaisha Efficient spectral imaging based on imaging systems with scene adaptation using tunable color pixels
US8908013B2 (en) * 2011-01-20 2014-12-09 Canon Kabushiki Kaisha Systems and methods for collaborative image capturing

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH04114622A (ja) * 1990-09-03 1992-04-15 Toshiba Corp 内視鏡用撮像素子モジュール
JP3920683B2 (ja) * 2002-03-28 2007-05-30 富士フイルム株式会社 撮像素子
JP6197291B2 (ja) * 2012-03-21 2017-09-20 株式会社リコー 複眼カメラ装置、及びそれを備えた車両
TWI518886B (zh) * 2012-04-24 2016-01-21 精材科技股份有限公司 光感裝置及其製法
CN105578078B (zh) * 2015-12-18 2018-01-19 广东欧珀移动通信有限公司 图像传感器、成像装置、移动终端及成像方法
CN106210532B (zh) * 2016-07-29 2019-10-11 宇龙计算机通信科技(深圳)有限公司 一种拍照处理方法及终端设备
CN107221544B (zh) * 2017-06-02 2023-09-26 展谱光电科技（上海）有限公司 多光谱摄像装置及其摄像方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20120127334A1 (en) * 2010-11-18 2012-05-24 Canon Kabushiki Kaisha Adaptive spatial sampling using an imaging assembly having a tunable spectral response
US8908013B2 (en) * 2011-01-20 2014-12-09 Canon Kabushiki Kaisha Systems and methods for collaborative image capturing
US8836808B2 (en) * 2011-04-19 2014-09-16 Canon Kabushiki Kaisha Adaptive color imaging by using an imaging assembly with tunable spectral sensitivities
US8854515B2 (en) * 2011-09-09 2014-10-07 Canon Kabushiki Kaisha Efficient spectral imaging based on imaging systems with scene adaptation using tunable color pixels
US20140293091A1 (en) * 2012-05-21 2014-10-02 Digimarc Corporation Sensor-synchronized spectrally-structured-light imaging
CN103491319A (zh) * 2012-06-11 2014-01-01 索尼公司 成像装置

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116913937A (zh) * 2023-08-25 2023-10-20 浙桂(杭州)半导体科技有限责任公司 图像传感器件及其光学调控方法
CN116913937B (zh) * 2023-08-25 2023-12-22 浙桂(杭州)半导体科技有限责任公司 图像传感器件及其光学调控方法

Also Published As

Publication number Publication date
CN109314746B (zh) 2021-08-10
CN109314746A (zh) 2019-02-05

Similar Documents

Publication Publication Date Title
CN105611136B (zh) 一种图像传感器以及电子设备
WO2015159651A1 (ja) 光検出装置および固体撮像装置並びにそれらの製造方法
WO2017076005A1 (zh) 数据发送方法、数据接收方法及相关设备、系统
EP2338274A2 (en) A monitor having integral camera and method of operating the same
CN102334333A (zh) 改善成像系统的景深
CN105763774A (zh) 一种摄像模组、电子设备以及图像采集方法
US11460666B2 (en) Imaging apparatus and method, and image processing apparatus and method
WO2023179527A1 (zh) 摄像模组、摄像模组的控制方法以及电子设备
US20180247961A1 (en) Image sensor and image capturing apparatus
CN107818765A (zh) 显示设备亮度控制方法、装置及移动智能终端和显示设备
CN103730477B (zh) 基于暗角补偿的图像传感器像素阵列及其形成方法
JPWO2019078340A1 (ja) 撮像装置および方法、並びに、画像処理装置および方法
WO2020155739A1 (zh) 图像传感器、从其获取图像数据的方法及摄像设备
WO2019119452A1 (zh) 图像传感器及其获取图像的方法、智能设备
CN103066086B (zh) 一种cmos图像传感器像素阵列及其制造方法
WO2016192437A1 (zh) 一种3d图像捕获装置、捕获方法及3d图像系统
CN103364926A (zh) 阵列式镜头模组
CN113419394B (zh) 一种镜头组件、摄像头及成像方法
CN203350517U (zh) 阵列式镜头模组
CN104079903B (zh) 高动态范围的彩色及灰度图像传感器
TW200533956A (en) Image sensing device of improving image quality and reducing color shift effect
WO2023179589A1 (zh) 摄像模组以及电子设备
US11159741B2 (en) Imaging device and method, image processing device and method, and imaging element
CN209017154U (zh) 一种摄像机
WO2016003106A2 (ko) 멀티 애퍼처 카메라의 센서 어레이

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17935214

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 17935214

Country of ref document: EP

Kind code of ref document: A1