A method of detecting a support structure in a navigational controller includes: controlling a depth sensor to capture a plurality of depth measurements corresponding to an area containing the support structure on a ground plane; generating, for each of the depth measurements, a projected depth measurement on the ground plane; selecting a boundary set of the projected depth measurements defining a boundary of the projection; selecting, based on angles between adjacent pairs of the boundary set of projected depth measurements, a subset of the projected depth measurements; generating a region of interest based on the subset of the projected depth measurements; retrieving a candidate subset of the depth measurements corresponding to the projected depth measurements in the region of interest; and generating a support structure plane definition based on the candidate subset of depth measurements.

A method of controlling light emitters of a mobile automation apparatus includes: controlling a depth sensor to capture a plurality of depth measurements corresponding to an area containing a support structure; obtaining a support structure plane definition; selecting a subset of the depth measurements; determining, based on the subset of depth measurements and the support structure plane, whether the subset of the depth measurements indicates the presence of a sensitive receptor; when the determination is affirmative, disabling the light emitters; and when the determination is negative, controlling (i) the light emitters to illuminate the support structure and (ii) a camera to capture an image of the support structure simultaneously with the illumination.

Methods and apparatus to monitor environments are disclosed. Example audience measurement devices disclosed herein include means for executing a first recognition analysis on three-dimensional data representative of a first object within a threshold distance from a three-dimensional sensor. Disclosed example audience measurement devices also include means for executing a second recognition analysis on two-dimensional data representative of a second object outside the threshold distance from the three-dimensional sensor. Disclosed example audience measurement devices further include means for combining a first detection of a first person provided by the first recognition analysis and a second detection of a second person provided by the second recognition analysis to generate a people count for an environment.

Embodiments of the present disclosure disclose a method and apparatus for recognizing a face. A specific embodiment of the method includes: acquiring at least two facial images of a to-be-recognized face under different illuminations using a near-infrared photographing device; generating at least one difference image based on a brightness difference between each two of the at least two facial images; determining a facial contour image of the to-be-recognized, face based on the at least one difference image; inputting the at least two facial images, the at least one difference image, and the facial contour image into a pre-trained real face prediction value calculation model to obtain a real face prediction value of the to-be-recognized face; and outputting prompt information for indicating successful recognition of a real face, in response to determining the obtained real face prediction value being greater than a preset threshold.

A lighting device for an inspection apparatus. The lighting device may include a hollow housing having an inner planar reflective surface and an opposing inner concave dome-shaped reflective surface. The lighting device may further include at least one light source disposed at the inner planar reflective surface. The hollow housing may include an opening configured to be a light outlet. An inspection apparatus including the lighting device and an imaging device coupled to the lighting device.

A system for acquiring information from an environment, comprising: a light source for generating at least one beam; a first optical setup for converting the at least one beam into a distorted light pattern projectable onto an environment; and a second optical setup for converting an original view returned from the environment and comprising the distorted light pattern deformed by at least one surface of the environment into a corrected image comprising a corrected pattern.

An operation method of a texture recognition device and a texture recognition device are provided. The texture recognition device includes a light source array and an image sensor array; the light source array includes a plurality of light sources; the image sensor array includes a plurality of image sensors, which are configured to receive light emitted from the light sources and reflected to the image sensors by a texture for a texture collection; the operation method includes: in a process of the texture collection performed by the image sensor array, lighting a first group of light sources continuously arranged in a first pattern at a first moment, so that the first group of light sources continuously arranged in the first pattern serve as a photosensitive light source for the image sensor array, in which a length-width ratio of a region occupied by the first pattern is larger than two.

A medicine inspection assistance device, an image processing device and an image processing method are provided that appropriately recognize identification information irrespective of whether the identification information is an engraved mark or a printed character. The image processing device that obtains a plurality of taken images of a medicine, performs a process of enhancing an engraved mark portion of the medicine based on at least one taken image among the taken images and generates a first enhanced image, performs a process of enhancing a printed character portion of the medicine based on at least one taken image among the taken images and generates a second enhanced image, collates an integrated image obtained by integrating the first enhanced image and the second enhanced image with each other, with a master image, and determines whether the medicine to be dispensed and the dispensed medicine are identical to each other or not.

A medicine inspection support device, an image processing device, an image processing method and a program are provided that appropriately recognize identification information irrespective of whether the identification information is an engraved mark or a printed character. The image processing device obtains a plurality of taken images of a second medicine with light illumination directions to the surface of the second medicine different from each other, generates a first enhanced image by performing a process of enhancing an engraved mark portion of the second medicine, generates a second enhanced image by performing a process of enhancing a printed character portion of the second medicine, and collates the image designated by the designation information designating one of the first enhanced image and the second enhanced image as an image to be used for collation with the master image to determine whether the first medicine is identical to the second medicine.

A medicine inspection device, an image processing device, an image processing method, and a program are provided that appropriately extract identification information even when the positional relationship between a printed character and an engraved mark added to a medicine is not constant. The above problem is solved by an image processing device that performs a process of enhancing a printed character portion of a medicine based on at least one image taken with a surface of the medicine being irradiated with light to generate a printed-character-enhanced image, performs a process of extracting only an engraved mark portion from a three-dimensional shape of the medicine based on at least three images taken with illumination directions of the light to the surface of the medicine different from each other to generate an engraved-mark-extracted image, and generates a difference image between the printed-character-enhanced image and the engraved-mark-extracted image.