Abstract: Embodiments include a method for pruning anchor points from a feature map generated from a Light Detections And Ranging (LiDAR) point cloud, the method comprising: receiving, by a navigation system, a LiDAR point cloud from a LiDAR sensor, the LiDAR point cloud comprising data representing one or more objects in physical surroundings detected by the LiDAR sensor; extracting, by the navigation system, a feature map from the LiDAR point cloud, the feature map comprising a plurality of anchor points, each anchor point defined by an anchor box; smoothing, by the navigation system, the extracted feature map; determining, by the navigation system, density of pixels within the anchor box of each anchor point; and pruning, by the navigation system, anchor points from the feature map based on a plurality of factors related to the determined density of pixels within the box of each anchor point.

Abstract: There is provided a projection indication device that generates a projection image to be projected on parcel based on sensing information of the parcel, the device including: a processor; and a memory, in which by cooperating with the memory, the processor specifies parcel based on a distance image of the parcel included in sensing information, tracks the parcel based on a color image of the parcel included in the sensing information, and tracks the parcel based on the distance image of the parcel in a case where the color image of the parcel on which a projection image is projected includes a white region expressed in white which is not an original gradation.

Abstract: A facial image enhancement method includes: determining a plurality of feature regions in a facial image according to feature points of the facial image; correcting an initial enhancement weight coefficient of each feature region of the feature regions and obtaining a weight map of discrete enhancement on the facial image; obtaining a weight map of continuous enhancement on the facial image according to the weight map of discrete enhancement on the facial image; and performing enhancing the facial image according to the weight map of continuous enhancement on the facial image to obtain an enhanced facial image.

Abstract: The present invention discloses a noise processing method and apparatus. The method includes: selecting a pixel in an area array, which is formed by arranging a plurality of pixels in a matrix; obtaining pre-stored depth information corresponding to the pixel; performing pixel dispersion calculation according to the depth information corresponding to the pixel, to obtain at least one dispersion of the pixel, the dispersion being used to indicate a degree of dispersion between the pixel and another pixel in the area array; and identifying the pixel as a noise and removing the pixel from the area array when the at least one dispersion of the pixel is greater than a preset threshold.

Abstract: A dynamic image processing apparatus includes a hardware processor. The hardware processor extracts (i) a region of interest and/or (ii) a frame image of interest from a series of frame images obtained by dynamic imaging of a subject. Further, the hardware processor stores, of the series of the frame images, only (i) the extracted region of interest, (ii) the extracted frame image of interest or (iii) the extracted region of interest in the extracted frame image of interest in a storage.

Abstract: An ultrasound system with a deep learning neural network feature is used to eliminate haze artifacts in B mode images of the carotid artery by analysis of orthogonal information. In a described implementation the orthogonal information comprises the structural information of a B mode image and motion information of the same field of view as that of the B mode image. In another embodiment the neural network reduces haze artifacts by reducing TGC gain at the depth of artifacts.

Abstract: The disclosure provides an image processing device and method capable of removing a halo artifact and increasing contrast enhancement effect when enhancing the contrast of an image. The image processing method includes obtaining a first blurring image by performing interpolation based on a representative value of each of blocks, having a predetermined size, of a previous frame image; obtaining a second blurring image in which boundary information is restored, through a weighted sum of a current frame image and the first blurring image; and performing contrast enhancement on the current frame image by using a difference image between the second blurring image and the current frame image.

Abstract: An apparatus comprising at least one processor, and at least one memory including computer program code, wherein the at least one memory and the computer program code are configured, with the at least one processor, to: receive an input signal, the input signal comprising a piecewise-smooth signal and a noise signal; determine a prior probability density function of a representation of the piecewise-smooth signal, the representation of the piecewise-smooth signal comprising a combination of a smooth, Gaussian, variable and a jump, Cauchy, variable; determine a likelihood function based on the input signal; determine a posterior probability density function based on the likelihood function and the prior probability density function distribution; and estimate the piecewise-smooth signal based on an estimate of a posterior probability density function.

Abstract: A facial image enhancement method includes: determining a plurality of feature regions in a facial image according to feature points of the facial image; correcting an initial enhancement weight coefficient of each feature region of the feature regions and obtaining a weight map of discrete enhancement on the facial image; obtaining a weight map of continuous enhancement on the facial image according to the weight map of discrete enhancement on the facial image; and performing enhancing the facial image according to the weight map of continuous enhancement on the facial image to obtain an enhanced facial image.

Abstract: In a method and system for reconstructing computed tomography image data in which CT image data is de-noised. Then simulated noise is added, followed by another de-noising step to estimate the bias. Then, the estimated bias information is used to correct the original de-noised image data to arrive at second pass image data.

Abstract: A synthetically generated noise image is generated from at least one high signal-to-noise ratio target image and at least two zero-mean Gaussian noise images, scaled according to a noise scale. The images are combined in a non-linear manner to produce the synthetically generated noise image which can be used as a training image in a machine learning-based system that “denoises” images. The process can be repeated for a number of different noise scales to produce a set of training images. In one embodiment, the synthetically generated noise image IN is generated according to: IN=?{square root over (I12+I22)} where I is the original target image, I1=I+pG1 and I2=pG2, and where G1 and G2 are zero-mean Gaussian noise images, and p is the noise scale.

Abstract: A biological image processing method is provided. The method acquires first time-frequency data from an image data; processing the first time-frequency data into the second time-frequency data using a filter module; and converting the second time-frequency data into a time domain signal. The filter module is obtained by machine learning, and is trained using a first sample time-frequency signal as input and a second sample time-frequency signal as output, wherein the noise of the time domain signal corresponding to the second sample time-frequency signal is less than the noise of the time domain signal corresponding to the first sample time-frequency signal. A biological information detection device is also provided.

Abstract: An image processing method, an image processing system and a storage medium are provided. The image processing method includes: acquiring an initial image; extracting a contour of a predetermined target included in the initial image during a transmission process of pixel data of the initial image; and obtaining information of the predetermined target according to the contour of the predetermined target included in the initial image.

Abstract: A method for processing data in a video sequence comprising impulse noise (“salt-and-pepper”, “snow”, or other type), comprising, for filtering the noise, an application of a recursive filter (S2), given by: z(n)=z(n?1)+? if y(n)>z(n?1) z(n)=z(n?1)?? if y(n)<z(n?1) and z(n)=z(n?1) if y(n)=z(n?1) where: y(n) designates an element in the nth image of the succession, not processed by application of the sign filter; z(n?1) designates an element having a position corresponding to y(n), in the (n?1)th image of the succession, and processed by application of the sign filter; z(n) designates an element having a position corresponding to y(n), in the nth image of the succession, and processed by application of said sign filter, and ? is a strictly positive coefficient.

Abstract: A system includes a sensor including a set of cameras configured to capture a set of images of an enclosed space. The system further includes a processor and a non-transitory computer-readable medium containing instructions which, when executed on the processor, cause the processor to perform operations including: (a) receiving the set of images captured by the set of cameras; (b) pre-processing the set of images to increase quality and reduce effect of ambient lighting; (c) building a disparity map for the set of images based at least in part on a relationship between the set of cameras; and (d) determining an occupancy of the enclosed space based at least in part on the disparity map. The system can be used to determine occupancy in a trailer with irregularly shaped cargo using low cost sensors.

Abstract: In order to facilitate setting of an appropriate detection condition in detecting an object included in an image, an image processing apparatus obtains an image captured by an image capturing unit; causes a display unit to display first information representing the number of persons who have a size included in a first range that is a range from a first size to a second size; and accepts an operation of changing the first range by a user. The image processing apparatus causes the display unit to display the first information and second information representing the number of persons who have a size included in a second range that is a range based on the operation by the user.

Abstract: An optical sensor system mounted in a wheel arch of a car for determining trajectory of the car includes: a first optical sensor mounted in the wheel arch above a wheel and located behind a first clear casing that does not touch the wheel; and a second optical sensor mounted in the wheel arch on one side of the wheel and located behind a second clear casing that does not touch the wheel. The first optical sensor and second optical sensor perform a plurality of counts corresponding to respectively capturing a plurality of images of the wheel. The captured images are compared with a reference image to determine a 2D displacement of the wheel from its original position. The trajectory of the car is determined by calculating a turning degree of the wheel according to a trigonometric manipulation of the measured 2D displacement.

Abstract: A method of determining volumetric data of a predetermined anatomical feature is described. The method comprising determining volumetric data of one or more anatomical features present in a field of view of a depth sensing camera apparatus, identifying a predetermined anatomical feature as being present in the field of view of the depth sensing camera apparatus, associating the volumetric data of one of the one or more anatomical features with the identified predetermined anatomical feature, and outputting the volumetric data of the predetermined anatomical feature. An apparatus is also described.

Abstract: A method for a deconvolution of a digital input image (I(xi)) having a plurality of input voxels (xi), in particular a digital input image obtained from a medical observation device, such as a microscope or endoscope and/or using fluorescence, includes computing a local signal-to-noise ratio (SNR(xi)) within an input region (R(xi)) of the digital input image, the input region consisting of a subset of the plurality of input voxels of the digital input image and surrounding the current input voxel. A noise component (?(SNR)) is computed from the local signal-to-noise ratio, the noise component representing image noise (([h*f](xi), n(xi)) in the deconvolution. The noise component is limited to a predetermined minimum noise value (?min) for a local signal-to-noise ratio above a predetermined upper SNR threshold value (SNRmax) and is limited to a predetermined maximum noise value (?max) for a local signal-to-noise ratio below a predetermined lower SNR threshold value (SNRmin).

Abstract: A distance measurement device including a pixel array and a cover layer is provided. The cover layer is covered on the pixel array. The cover layer includes a first cover pattern covering on a first area of a plurality of first pixels and a second cover pattern covering on a second area of a plurality of second pixels. The first area and the second area are rectangles of mirror symmetry along a first direction.