THIN MULTI-APERTURE IMAGING SYSTEM WITH AUTO-FOCUS AND METHODS FOR USING SAME

Abstract

Dual-aperture digital cameras with auto-focus (AF) and related methods for obtaining a focused and, optionally optically stabilized color image of an object or scene. A dual-aperture camera includes a first sub-camera having a first optics bloc and a color image sensor for providing a color image, a second sub-camera having a second optics bloc and a clear image sensor for providing a luminance image, the first and second sub-cameras having substantially the same field of view, an AF mechanism coupled mechanically at least to the first optics bloc, and a camera controller coupled to the AF mechanism and to the two image sensors and configured to control the AF mechanism, to calculate a scaling difference and a sharpness difference between the color and luminance images, the scaling and sharpness differences being due to the AF mechanism, and to process the color and luminance images into a fused color image using the calculated differences.

Claims

1. A dual-aperture digital camera with autofocus (AF) for imaging an object or scene, comprising: a) a first sub-camera that includes a first optics bloc and a first, color image sensor with a first number of pixels, the first camera operative to provide a first, color image of the object or scene; b) a second sub-camera that includes a second optics bloc and a second image sensor having a second number of pixels, the second sub-camera operative to provide a second image of the object or scene; c) an AF mechanism coupled mechanically at least to the first optics bloc and used to perform an AF action on at least the first optics bloc; and d) a camera controller coupled to the AF mechanism and to the two image sensors and configured, based on calibration data acquired through a calibration step, to control the AF mechanism, to estimate the scale difference between the first and second images and to process the first and second images into a fused color image using the calculated scale difference.

2. The dual-aperture digital camera of claim 1, wherein the color image sensor includes a non-standard color filter array.

3. The dual-aperture digital camera of claim 1, wherein the camera controller configuration is based on calibration data acquired through a calibration step.

4. The dual-aperture digital camera of claim 3, wherein the calibration data is saved in a one-time programmable memory or an EEPROM in the dual-aperture digital camera.

5. The dual-aperture digital camera of claim 1, wherein the first number of pixels and the second numbers of pixels are the same.

6. The dual-aperture digital camera of claim 1, wherein the first number of pixels and the second numbers of pixels are different.

7. The dual-aperture digital camera of claim 1, wherein the first, color image sensor and the second image sensor have pixels of the same size.

8. The dual-aperture digital camera of claim 1, wherein the camera controller is further configured to find corresponding points in the first and second images, estimate a scale factor and use the scale factor to process the first and second images into a fused color image.

9. The dual-aperture digital camera of claim 1, wherein the camera controller is further configured to preprocess the first and second images, to obtain respective rectified first and second images, to register the rectified first and second images to obtain respective registered first and second images, and to fuse the registered first and second images into a fused color image.

10. The dual-aperture camera of claim 1, wherein the first sub-camera includes an infra-red (IR) filter that blocks IR wavelengths from entering the first image sensor and wherein the second sub-camera is configured to allow at least some IR wavelengths to enter the second image sensor.

11. The dual-aperture camera of claim 1, wherein the second image sensor is a luminance image sensor and wherein the second image is a luminance image.

12. The dual-aperture digital camera of claim 8, wherein the camera controller is further configured to preprocess the first and second images, to obtain respective rectified first and second images, to register the rectified first and second images to obtain respective registered first and second images, and to fuse the registered first and second images into a fused color image.

13. The dual-aperture digital camera of claim 9, wherein epipolar lines in the rectified first and second images are more-or-less parallel to horizontal or vertical axes of the rectified first and second images.

14. The dual-aperture digital camera of claim 12, wherein epipolar lines in the rectified first and second images are more-or-less parallel to horizontal or vertical axes of the rectified first and second images.

15. The dual-aperture digital camera of claim 14, wherein the camera controller configuration is based on calibration data acquired through a calibration step, wherein the calibration data is saved in a one-time programmable memory or an EEPROM in the dual-aperture digital camera.

16. The dual-aperture digital camera of claim 15, wherein the first number of pixels and the second numbers of pixels are different.

17. A method, comprising: in a dual camera: a) obtaining simultaneously a first, color image of an object or scene with a first sub-camera and a second image of the object or scene with a second sub-camera, wherein the first sub-camera includes a first optics bloc, wherein the second sub-camera includes a second optics bloc and wherein the first and second sub-cameras have substantially the same field of view; b) using an AF mechanism coupled mechanically at least to the first optics bloc to perform an AF action on at least the first optics bloc; c) preprocessing the first image and the second image to obtain respective rectified and scale-adjusted first and second images while considering scaling differences caused by the AF action; d) performing registration between the rectified and scale-adjusted first and second images to obtain registered images; and e) fusing the registered images into a focused fused color image.

18. The method of claim 17, wherein the fusing of the registered images is not performed in non-focused areas.

19. The method of claim 17, wherein the preprocessing to obtain respective rectified scale-adjusted first and second images includes calculating a set of corresponding points in the first and second images and estimating a scale factor S.

20. The method of claim 19, further comprising using scaling factor S to scale one of the images to match the other image, thereby obtaining the registered images.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Non-limiting examples of embodiments disclosed herein are described below with reference to figures attached hereto that are listed following this paragraph. The drawings and descriptions are meant to illuminate and clarify embodiments disclosed herein, and should not be considered limiting in any way.

[0022] FIG. 1A shows schematically the design of a traditional digital camera;

[0023] FIG. 1B shows schematically the design of a dual-aperture camera;

[0024] FIG. 2 shows schematically an embodiment of a dual-aperture imaging system with auto-focus disclosed herein, in (a) a general isomeric view, and (b) a sectioned isomeric view;

[0025] FIG. 3 shows an embodiment of an image sensor for the imaging system in FIG. 2, in which one sub-camera has a CFA sensor, while another sub-camera has a clear sensor;

[0026] FIG. 4A shows schematically in a flow chart an embodiment of a method disclosed herein;

[0027] FIG. 4B shows in a flow chart details of the scale adjustment step in the method shown in FIG. 4A;

[0028] FIG. 4C shows two images with corresponding points;

[0029] FIG. 5A shows schematically another embodiment of a dual-aperture imaging system with a single auto-focus mechanism disclosed herein in a sectioned isomeric view;

[0030] FIG. 5B shows schematically in a flow chart an embodiment of a method for auto-focus imaging with the imaging system in FIG. 5A;

[0031] FIG. 6 shows schematically yet another embodiment of a dual-aperture imaging system numbered with a single auto-focus mechanism in a sectioned isomeric view.

DETAILED DESCRIPTION

[0032] FIG. 2 shows schematically an embodiment of a dual-aperture imaging system with auto-focus disclosed herein and numbered 200, in (a) a general isomeric view, and (b) a sectioned isomeric view. In the following description, imaging system and camera may be used interchangeably. System 200 comprises two sub-cameras, labeled 202 and 204, each sub-camera having its own optics. Thus, sub-camera 202 includes an optics bloc 206 with an aperture 208 and an optical lens module 210, as well as a sensor 212. Similarly, sub-camera 204 includes an optics bloc 214 with an aperture 216 and an optical lens module 218, as well as a sensor 220. The sensors are also referred to henceforth as sensor 1 (212) and sensor 2 (220). Note that the two sensors may be implemented as two distinct areas on the same substrate, and not necessarily as two stand-alone sensors. Each optical lens module may include several lens elements as well as an Infra-Red (IR) filter 222a, b. In some embodiments, some or all of the lens elements belonging to different apertures may be formed on the same substrate. The two sub-cameras are positioned next to each other, with a small baseline 224 between the two apertures 208 and 216. Each sub-camera further includes an auto-focus mechanism, respectively 226 and 228.

[0033] The sensors used in each sub-camera may have different color filter arrays (CFAs). In some embodiments, sensor 1 may have one type of CFA, while sensor 2 may have another type of CFA. In some embodiments, sensor 1 may have a CFA and sensor 2 may have a white or clear filter array (marked by W)in which all the pixels absorb the same wide range of wavelengths, e.g. between 400 nm and 700 nm (instead of each pixel absorbing a smaller portion of the spectrum). A sensor having a color filter array may be referred to henceforth as a color image sensor, while a sensor with a clear or W filter array is referred to as a clear image sensor. FIG. 3A shows a sensor embodiment 300, where numeral 1 represents sensor 1 (with a CFA) and numeral 2 represents sensor 2 (with a clear white filter array). Circles 302a, 302b mark image circles formed by the optics on the sensors, while a white area 304 marks the substrate on which the sensors are located. Circles 302a, 302b may be larger than the respective size of the sensor the image is formed on. In some cases, overlap between the two image circles may occur and mechanical light blocking elements (e.g., walls) may be needed to prevent optical cross-talk between the sub-cameras.

[0034] The CFA of sensor 1 may be standard or non-standard. As used herein, a standard CFA may include a known CFA such as Bayer, RGBE, CYYM, CYGM and different RGBW filters such as RGBW#1, RGBW#2 and RGBW#3. For example, non-Bayer CFA patterns include repetitions of a 2?2 micro-cell in which the color filter order is RRBB, RBBR or YCCY where Y=Yellow=Green+Red, C=Cyan=Green+Blue; repetition of a 3?3 micro-cell in which the color filter order is GBRRGBBRG (e.g. as in sensor 1 in FIG. 3A); and repetitions of a 6?6 micro-cell in which the color filter order is one of the following options: [0035] 1. Line 1: RBBRRB. Line 2: RWRBWB. Line 3: BBRBRR. Line 4: RRBRBB. Line 5: BWBRWR. Line 6: BRRBBR. [0036] 2. Line 1: BBGRRG. Line 2: RGRBGB. Line 3: GBRGRB. Line 4: RRGBBG. Line 5: BGBRGR. Line 6: GRBGBR. [0037] 3. Line 1: RBBRRB. Line 2: RGRBGB. Line 3: BBRBRR. Line 4: RRBRBB. Line 5: BGBRGR. Line 6: BRRBBR. [0038] 4. Line 1: RBRBRB. Line 2: BGBRGR. Line 3: RBRBRB. Line 4: BRBRBR. Line 5: RGRBGB. Line 6: BRBRBR.

[0039] The color CFA of sensor 1 in FIG. 3B is a Bayer pattern. By using a CFA on sensor 1, sub-camera 1 captures the color information about the scene, while sub-camera 2 captures luminance information about the scene.

[0040] The CFA pattern of sensor 1 in FIG. 3A as well as other non-Bayer CFAs listed above an advantage over the standard Bayer pattern in that they divide the red, green and blue colors evenly across the sensor pixels. This results in a finer sampling of red and blue colors, while the green color experiences coarser sampling compared with the standard Bayer pattern. However, as the image that is captured by sensor 2 is used to extract luminance information about the scene (instead of relying on the green channel for that, as is the case when using a Bayer CFA), the green pixels are only used for color information. In traditional compact camera design, a filter that lets in light in the visible range and blocks light in the IR range is typically placed in the optical path, sometimes as part of a cover glass that protects the sensor. Although the blocking of IR light wastes photons, it allows for a more accurate estimation of the color in the scene, as it reduces color crosstalk from the spectral response of the R, G and B color filters (which may be sensitive to IR light). In an embodiment, clear sensor 2 is made sensitive to IR light by removing the IR filter or by redesigning its spectral response to let in some light in the IR range. The motivation for capturing IR light, in addition to light in the visible range, is to increase the Signal-to-Noise Ratio (SNR) in the image, as many natural and artificial light sources also emit photons in the IR spectrum.

[0041] Unlike a sensor with a color CFA (i.e. sensor 1), absorption of IR light does not introduce color cross-talk in clear sensor 2 (since the sensor records a panchromatic image of the scene).

[0042] Removing the IR filter may have some negative implications on image quality. For example, extending the range of wavelengths that are captured by the camera may lead to longitudinal chromatic aberrations that may degrade the Point Spread Function (PSF), resulting in a blurrier image. To address this issue, in an embodiment, the optics of sub-camera 2 are optimized across both the visible and the IR range, to mitigate the effect of chromatic aberrations and to result in a more compact PSF compared with standard compact camera optics that use an IR filter. This is unlike the standard optimization process, which considers only wavelengths inside the visible range.

[0043] In use, the two sub-cameras share a similar FOV and have substantially equal (limited only by manufacturing tolerances) focal lengths. An image capture process is synchronized, so that the two sub-cameras capture an image of the scene at a particular moment. Due to the small baseline between the two apertures (which could be only a few millimeters, for example 6.5 mm or 8.5 mm) of the sub-cameras, the output images may show parallax, depending on the object distances in the scene. A digital image processing algorithm combines the two images into one image, in a process called image fusion. Henceforth, the algorithm performing this process is called image fusion algorithm. The resulting image may have a higher resolution (in terms of image pixels) and/or a higher effective resolution (in terms of the ability to resolve spatial frequencies in the scene, higher effective resolution meaning the ability to resolve higher spatial frequencies) and/or a higher SNR than that of one sub-camera image.

[0044] In terms of resolution and exemplarily, if each sub-camera produces a 5 megapixel (2592?1944 pixels) image, the image fusion algorithm may combine the two images to produce one image with 8 megapixel (3264?2448 pixels) resolution. In terms of effective resolution, assuming that an imaged object or scene includes spatial frequencies, the use of a dual-aperture camera having a clear sensor and a color sensor as disclosed herein leads to an overall increase in effective resolution because of the ability of the clear sensor to resolve higher spatial frequencies of the luminance component of the scene, compared with a color sensor. The fusion of the color and clear images as performed in a method disclosed herein (see below) adds information in spatial frequencies which are higher than what could be captured by a color (e.g. Bayer) sub-camera.

[0045] In order to generate a higher-resolution or higher effective resolution image, the image fusion algorithm combines the color information from sub-camera 1 with the luminance information from sub-camera 2. Since clear sensor 2 samples the scene at a higher effective spatial sampling rate compared with any color channel or luminance thereof in the color sensor 1, the algorithm synthesizes an image that includes information at higher spatial frequencies compared with the output image from sub-camera 1 alone. The target of the algorithm is to achieve a spatial resolution similar to that obtained from a single-aperture camera with a sensor that has a higher number of pixels. Continuing the example above, the algorithm may combine two 5 megapixel images, one color and one luminance, to produce one 8 megapixel image with information content similar to that of a single-aperture 8 megapixel color camera.

[0046] In addition to improved spatial resolution, the image fusion algorithm uses the luminance information from clear sensor 2 to generate an image with increased SNR, vs. an image from a corresponding single-aperture camera. The fact that the pixels of sensor 2 are not covered by color filters allow each pixel to absorb light in a wider wavelength spectrum, resulting in a significant increase in the light efficiency compared with a color CFA camera. In an embodiment, the fusion of clear image information and color image information then provides a +3 dB SNR increase over that of a single aperture digital camera.

[0047] As clear sensor 2 is more sensitive than color sensor 1, there may be a need to adjust exposure times or analog gains to match the digital signal levels between the two cameras. This could be achieved by fixing the same exposure times to both sensors and configuring a different analog gain to each sensor, or by fixing the analog gain in both sensors and configuring a different exposure time to each sensor.

[0048] FIG. 4A shows schematically, in a flow chart, an embodiment of a method disclosed herein. FIG. 4B shows in a flow chart details of the scale adjustment step in the method shown in FIG. 4A. Two images 400a and 400b from respectively sub-cameras 1 and 2 serve as inputs. The two images undergo pre-processing, in respectively step 402a for the color image of sensor 1 and 402b for the luminance image of sensor 2. Step 402a includes digital image signal processing (ISP) in an ISP pipeline. The ISP generates a full color image, with R, G, B values at each image pixel. If the CFA pattern on sensor 1 is non-Bayer, the ISP includes non-standard demosaicing to interpolate the missing colors at each pixel location. In addition to demosaicing, other standard ISP pipeline algorithms may be applied on the image, e.g., black level correction, defect pixel correction, noise removal, etc, as known in the art. The luminance image from sub-camera 2 is also pre-processed to correct for defects, noise, shading profile, blur and other optical, analog and digital aberrations. Normalization, rectification and scale adjustment are then applied on the two images in step 404. First, the two images are normalized to have the same mean signal intensity and standard deviation (which is a measure for the image dynamic range). This is done by subtracting the mean from each pixel and dividing each pixel by the standard deviation in each image. Then, the images are rectified by applying two projection matrices, in order to correct for different rotations around the x, y and z axes, to correct for x-y translations of the optical center of the two cameras and to fix lens distortions. The projection matrices parameters are pre-calculated from calibration data, which may be acquired through a calibration step that is applied for each camera module during camera module assembly. The data may be saved in one-time programmable memory or EEPROM in the camera module. After the rectification step, epipolar lines in both images are more-or-less parallel to the horizontal axis of the image, in case the two sub-cameras are positioned one beside the other along the X-axis, or parallel to the vertical axis of the image, in case the two sub-cameras are positioned one beside the other along the Y axis.

[0049] The scale adjustment, done after the rectification step, is described now in more detail with reference to FIGS. 4B. Preprocessed and rectified images 418a and 418b (also shown exemplarily in FIG. 4C) from respectively sub-cameras 1 and 2 serve as inputs. In step 420, corresponding points between the two images are found. In an embodiment, the set of corresponding points is calculated over the entire image. In another embodiment, the set of corresponding points is found for a specific region of interest (ROI) in each image. FIG. 4C, which shows schematically two images A and B of the same scene captured by adjacent cameras (i.e. A captured by sub-camera 1 and A captured by sub-camera 2) with some parallaxdue to the different viewpoint, objects are imaged with some displacement in one image compared with the other, depending on their distance from the cameras. Pairs of features a-b-b and c-c represent the same corresponding points in the two images A and A. An algorithm is used to find corresponding points between the two images. A set of prominent points are found (e.g. corners) in the two images and then the algorithm finds matches between the points in the two images. Such algorithms are known to the skilled in the art. In step 422, the Y coordinate only is extracted in order to estimate the scale between the two images. Since the position of the optics, which is controlled by the AF mechanism, may introduce different scales between the two sub-camera images, the proper scale needs to be determined for each captured image (i.e. for each focus position). Assuming the two sub-cameras are positioned adjacent to one another along the X-axis, once corresponding pairs of points are found, a single coordinate is extracted from each point in step 422. That is, the algorithm considers only their Y coordinate and disregards their X coordinate. The inventors have advantageously realized that while the X coordinate may be affected by parallax, the Y coordinate is largely unaffected by parallax after the rectification step, and therefore the Y coordinates can be used to estimate the scale more robustly. If the two sub-cameras are positioned adjacent along the Y-axis, then once corresponding pairs of point are found, the algorithm considers only their X coordinate and disregards their Y coordinate. Continuing with the assumption of the two sub-cameras being adjacent along the X-axis, the Y coordinates of the corresponding points are used to estimate a scaling factor S between the images in step 424. In an exemplary embodiment, the scaling factor estimation is performed using least-squares, in which case S is given by

S=(Y2*W*Y2)\Y2*W*Y1

where Y1 is a vector of Y coordinates of points taken from one image, Y2 is a vector of Y coordinates of points taken from the other image, and W is a diagonal matrix that holds the absolute values of Y2. Scaling factor S is then used in step 426 to scale one image in order to match the scale between the two images. In step 426, point coordinates in each image are multiplied by the same scaling factor S. Finally, in step 428, the corresponding pairs of scaled points are used to calculate a shift in x and y axes between the two images for each axis. In an embodiment, only a subset of the corresponding points that lie in a certain ROI is used to calculate the shift in x and y. For example, the ROI may be the region used to determine the focus, and may be chosen by the user or the camera software (SW). The estimated shift is applied on one of the images or on both images. The result of the scale adjustment process in FIG. 4B (and in step 404, FIG. 4A) are scaled images 430.

[0050] Returning now to FIG. 4A, local registration and parallax correction to estimate a disparity map are applied to the scaled images in step 406. The local registration uses scale and shift parameters found in step 404. Fusion to enhance the resolution and improve SNR in the final image is then performed in step 408, by combining information from both images, according to the disparity map. The fusion process uses the image from sub-camera 1 as a baseline. The output is a fused image 410. Post-processing such as tone mapping, gamma correction, contrast enhancement and color correction/enhancement may then be applied to the fused image.

Auto-Focus

[0051] As mentioned with respect to FIG. 2, a camera system disclosed herein includes an AF mechanism that controls the focus position of the optics. The system shown in FIG. 2 includes two such AF mechanisms. FIG. 5A shows schematically another embodiment of a dual-aperture imaging system numbered 500 with a single auto-focus mechanism in a sectioned isomeric view. System 500 includes in addition to the regular image sensors and optics only one AF mechanism 502, positioned in a color sub-camera 1. A luminance sub-camera 2 does not have an AF mechanism, being instead a fixed-focus camera, with the focus fixed to a certain object distance. The focus position is such that the DOF range of sub-camera 2 is between infinity and several tens of centimeters, depending on the focal length and optical design. For example, the DOF may be between infinity and 50 cm, such that sub-camera 2 would produce sharp images for object distances that lie within this range from the camera. In system 500, sub-camera 1 can produce an image in which the main object is in focus for a wide range of object distances, so that it appears sharp in a sub-camera 1 image, by changing the focus position of the optics.

[0052] FIG. 5B shows schematically in a flow chart an embodiment of a method for image fusion using an imaging system 500 that has AF. Two images 500a and 500b from respectively sub-cameras 1 and 2 serve as inputs. A focus position is chosen for sub-camera 1 in step 502. A check is performed in step 504 to determine whether the distance of an imaged object lies within the DOF of sub-camera 2, by calculating a sharpness metric on the images of sub-cameras 1 and 2, as known in the art. The calculation of the sharpness metric may result in a sharpness difference. If the answer in the check of step 504 is Yes, the object will appear sharp in the sub-camera 2 image. In such a case, image fusion as described above is applied to the object image obtained by both sub-cameras in step 506 to achieve higher output resolution and better SNR. If the answer to check 504 is No (i.e. the object lies closer to the camera, outside the DOF range of sub-camera 2), the object will appear blurry (not sharp) in the sub-camera 2 image. In this case, the image from sub-camera 2 is not used to enhance the resolution, but only to improve the SNR of the image from sub-camera 1. To this end, another algorithm (procedure) similar to the fusion algorithm (procedure) above is applied in step 508. The image from sub-camera 1 is scaled to the proper output size and a de-noising algorithm that uses information from the sub-camera 2 image is applied. Since in this case high frequencies are lost in the sub-camera 2 image (due to defocus), the algorithm only considers information at low spatial frequencies from the image of sub-camera 2. In order to determine the object distance, the chosen focus position of the AF mechanism of sub-camera 1 is used (after the focusing process has converged).

[0053] FIG. 6 shows schematically yet another embodiment of a dual-aperture imaging system numbered 600 with a single AF mechanism in a sectioned isomeric view. Similar to system 500, system 600 includes in addition to the regular image sensors and optics only one AF mechanism 602. However, in contrast with AF mechanism 502, AF mechanism 602 moves the optics of sub-camera 1 and the optics of sub-camera 2 together. The optical elements are mounted on a lens holder 604 with dedicated threads to hold the two lenses, which is moved by the AF mechanism. Since the optics of sub-camera 1 and sub-camera 2 have very similar focal lengths, the mechanical movement brings the image from sub-camera 1 and from sub-camera 2 to focus at the same time. The advantage of this construction over having only one AF mechanism is that both sub-cameras support the same range of object distances, so that the image fusion algorithm can be applied for the entire range. When the AF mechanism chooses the best focus position for the lens, information from both sub-camera images can be taken into account (e.g. to assist in focusing in low-light situations). In low-light, AF sharpness measurements are noisier, due to the lower SNR in the images.

[0054] Using two images instead of one can help reduce the noise and improve the robustness and accuracy of the AF process (algorithm).

[0055] In an embodiment, some or all the optical elements of sub-camera 1 and sub-camera 2, are made on the same die, using wafer-level optics manufacturing techniques or injection molding of glass or plastic materials. In this case, the single AF mechanism moves the optical dies on which the optical elements of the two sub-cameras are fabricated, so that the two optical stacks move together.

[0056] In another embodiment, a camera is similar to camera 500 and includes a single AF mechanism placed on sub-camera 1 (with the color CFA). Sub-camera 2 does not have an AF mechanism, but uses instead fixed focus optics with unique characteristics that provide extended depth of focus, which is achieved by means of optical design (e.g., by employing optics with narrower aperture and higher F-number). The optical performance of the optics of sub-camera 2 is designed to support sharp images for object distances between infinity and several cm from the camerain this case, the fusion algorithm can be applied to enhance output resolution for a wider range of object distances compared with the single AF embodiment described above. There is usually a tradeoff between the DOF of the camera and the minimal achievable PSF size across the DOF range. An algorithm may be used to enhance the sharpness of the image captured by sub-camera 2 before the fusion algorithm is applied to combine the photos. Such an algorithm is known in the art.

[0057] To conclude, dual-aperture cameras and methods of use of such cameras disclosed herein have a number of advantages over single aperture cameras, in terms of camera height resolution, effective resolution and SNR. In terms of camera height, in one example, a standard 8 Mpix ? camera with a 70 degree diagonal FOV may have a module height of 5.7 mm. In comparison, a dual-aperture camera disclosed herein, with two 5 Mpix ? image sensors (one color and one clear), each with 70 degrees diagonal field of view may have a module height of 4.5 mm In another example, a standard 8 Mpix ? camera with a 76 degree diagonal FOV may have a module height of 5.2 mm. In comparison, a dual-aperture camera disclosed herein, with two 5 Mpix ? image sensors (one color and one clear), each with a 76 degree diagonal FOV, may have a module height of 4.1 mm.

[0058] While this disclosure has been described in terms of certain embodiments and generally associated methods, alterations and permutations of the embodiments and methods will be apparent to those skilled in the art. The disclosure is to be understood as not limited by the specific embodiments described herein, but only by the scope of the appended claims