An optical system includes a plurality of internal apertures, a plurality of external optical assemblies and a telescope assembly positioned between the plurality of internal apertures and the plurality of external optical assemblies. Each internal aperture is operable to receive a corresponding aperture-specific optical signal. Each external optical assembly corresponds to one of the internal apertures, and each external optical assembly is operable to direct the aperture-specific optical signal of the corresponding internal aperture in a corresponding external direction. The external direction for each external optical assembly is independently controllable and the telescope assembly defines a shared optical train arranged to direct the aperture-specific optical signals between each internal aperture and the corresponding external optical assembly.

The present disclosure relates to an image-capturing apparatus, an image-capturing method, and an image-capturing device that allow reduction of an influence of diffraction due to opening sections of a mask in lensless image-capturing. Before a mask, a band-pass filter that is divided into a plurality of areas, each of which transmits incident light in a different wavelength band, is provided, and the mask that includes opening sections, and modulates the incident light, which has been transmitted through the band-pass filter, in the wavelength bands that are different for the individual areas is provided. The opening sections of the mask have such unit sizes that blurring resulting from diffraction which occurs to each wavelength of the incident light is minimized. The present disclosure can be applied to a lensless image-capturing apparatus.

A spatial filtering apparatus includes a composite filter including first filter patterns respectively having a first phase profile, and second filter patterns respectively having a second phase profile, wherein the first filter patterns and the second filter patterns overlap with each other, wherein first light in a first polarization direction that is emitted on the composite filter is first spatially filtered by the first filter patterns, and wherein second light in a second polarization direction that is emitted on the composite filter is second spatially filtered by the second filter patterns.

A spatial filtering apparatus includes a composite filter including first filter patterns respectively having a first phase profile, and second filter patterns respectively having a second phase profile, wherein the first filter patterns and the second filter patterns overlap with each other, wherein first light in a first polarization direction that is emitted on the composite filter is first spatially filtered by the first filter patterns, and wherein second light in a second polarization direction that is emitted on the composite filter is second spatially filtered by the second filter patterns.

An imaging apparatus includes an imaging device, a first imaging optical system and a second imaging optical system that form respective input images from mutually different viewpoints onto the imaging device, and a first modulation mask and a second modulation mask that modulate the input images formed by the first imaging optical system and the second imaging optical system. The imaging device captures a superposed image composed of the two input images that have been formed by the first imaging optical system and the second imaging optical system, modulated by the first modulation mask and the second modulation mask, and optically superposed on each other, and the first modulation mask and the second modulation mask have mutually different optical transmittance distribution characteristics.

An imaging apparatus includes an imaging device, a first imaging optical system and a second imaging optical system that form respective input images from mutually different viewpoints onto the imaging device, and a first modulation mask and a second modulation mask that modulate the input images formed by the first imaging optical system and the second imaging optical system. The imaging device captures a superposed image composed of the two input images that have been formed by the first imaging optical system and the second imaging optical system, modulated by the first modulation mask and the second modulation mask, and optically superposed on each other, and the first modulation mask and the second modulation mask have mutually different optical transmittance distribution characteristics.

Aspects of the present invention provide a new text location technique, which can be applied to general handwriting detection at a variety of levels, including characters, words, and sentences. The inventive technique is efficient in training deep learning systems to locate text. The technique works for different languages, for text in different orientations, and for overlapping text. In one aspect, the technique's ability to separate overlapping text also makes the technique useful in application to overlapping objects. Embodiments take advantage of a so-called skyline appearance that text tends to have. Recognizing a skyline appearance for text can facilitate the proper identification of bounding boxes for the text. Even in the case of overlapping text, discernment of a skyline appearance for words can help with the proper identification of bounding boxes for each of the overlapping text words/phrases, thereby facilitating the separation of the text for purposes of recognition.

Aspects of the present invention provide a new text location technique, which can be applied to general handwriting detection at a variety of levels, including characters, words, and sentences. The inventive technique is efficient in training deep learning systems to locate text. The technique works for different languages, for text in different orientations, and for overlapping text. In one aspect, the technique's ability to separate overlapping text also makes the technique useful in application to overlapping objects. Embodiments take advantage of a so-called skyline appearance that text tends to have. Recognizing a skyline appearance for text can facilitate the proper identification of bounding boxes for the text. Even in the case of overlapping text, discernment of a skyline appearance for words can help with the proper identification of bounding boxes for each of the overlapping text words/phrases, thereby facilitating the separation of the text for purposes of recognition.

Embodiments describe optical imagers that include one or more micro-optic components. Some imagers can be passive imagers that include a light detection system for receiving ambient light from a field. Some imagers can be active imagers that include a light emission system in addition to the light detection system. The light emission system can be configured to emit light into the field such that emitted light is reflected off surfaces of an object in the field and received by the light detection system. In some embodiments, the light detection system and/or the light emission system includes micro-optic components for improving operational performance.

A 3D image sensor based on standard Complementary Metal Oxide Semiconductor (CMOS) process is described. The conventional CMOS image sensor measures a 2D projection of the 3D world in gray-scale or color image; this new sensor can measure the third dimension on the 2D image—the depth of object on the 2D image pixels. Since the standard CMOS image sensor can only sense intensity (the number of photons) at each CMOS pixel, this new sensor creates a new mechanism to encode the depth information into an intensity distribution change that CMOS sensor can sense. The idea is based on the observation or lens imaging theory that light cone for any point (pixel) on the image plane is narrower for a near object and wider for a distant object. We then use diffraction to measure the change of incident angle, based on the theory that an oblique light goes through a finite grating producing diffraction patterns multiple times from near field to far field, the incident angle is reflected as diffraction pattern shift. We use a normal CMOS patterning process to create gratings on top of photosensitive material, and place multiple pixels at a certain distance from the gratings to sense the intensity distribution change or shift for different light cones. Then the depth can be calculated by solving the imaging inverse problem. The solution or intermedium solution of such an inverse problem can be pre-calculated or pre-calibrated and placed in lookup tables so that the image sensor can output directly depth information from the 3D CMOS image sensor.