Abstract: The disclosure relates to a method for training a neural network classifier (100) to classify a digital image depicting one or more objects of a biological sample into a specific class, which class belongs to a predefined set of classes (C1-C3), the method comprising: providing a training set of digital images (110a-s) originating 5 from a plurality of biological samples, each digital image (110a) of the training set being labeled with a specific class (C1) of the one or more objects of the digital image (110a), each digital image (110a) of the training set being associated with global data (114a) pertaining to the respective sample, training the neural network (100) using pixel data of each digital image (110a) from the training set of digital 10 images (110a-s) and the global data (114a) associated with said digital image (110a) as input, and using the specific class (C1) of the label of said digital image (110a) as a correct output from the neural network (100).

Abstract: The disclosure relates to a method and corresponding system for digital microscopy imaging comprising: capturing with a camera, a plurality of images of a position in a biological sample, wherein each image is captured at a different focal distance, and determining in the camera, a focus value for each captured image. For each captured image, the focus value of the captured image is compared with at least one threshold focus value. Upon determining that the focus value of the captured image exceeds the at least one threshold focus value, the captured image is marked as an interesting image, and transmitted from the camera to a separate computing unit.

Abstract: In one embodiment of the present invention, a method is disclosed for determining a difference between a sample position and an in-focus position, as well as a vision inspection system. In a first step image data depicting a sample is captured. Next, a feature set is extracted from the image data. Thereafter, the feature set is classified into a position difference value, corresponding to the difference between the sample position and the in-focus position, by using a machine learning algorithm that is trained to associate image data features to a position difference value.

Abstract: Methods and systems for generating a graphical user interface for analysis of red blood cells. The method generates a first view of the graphical user interface by highlighting red blood cells in an image that are classified into at least one group indicated by a user input. The method also generates a second view by extracting individual red blood cells being classified into the at least one group indicated by the user input from the image of red blood cells and displaying them group-wise.

Abstract: Methods and systems for generating a graphical user interface for analysis of red blood cells. The method generates a first view of the graphical user interface by highlighting red blood cells in an image that are classified into at least one group indicated by a user input. The method also generates a second view by extracting individual red blood cells being classified into the at least one group indicated by the user input from the image of red blood cells and displaying them group-wise.

Abstract: In one embodiment of the present invention, a method is disclosed for determining a difference between a sample position and an in-focus position, as well as a vision inspection system. In a first step image data depicting a sample is captured. Next, a feature set is extracted from the image data. Thereafter, the feature set is classified into a position difference value, corresponding to the difference between the sample position and the in-focus position, by using a machine learning algorithm that is trained to associate image data features to a position difference value.

Abstract: An image processing system for optical systems including microscopes which separately captures (grabs) different portions of the electromagnetic spectrum as emitted from an object as separate images. Each of the separate images are individually focused and then aligned with the other images to produce a composite image. In one embodiment, the different portions correspond to red, green, and blue and the composite image is displayed to the user on as a monitor without any need for an eyepiece.

Abstract: An image registration (alignment) system which achieves alignment by determining an optimal focus for a composite image produced from the combination of misregistered images. The system also uses optimal or nearly optimal linear filters for filtering the combined image to emphasize (enhance) alignment information in the combined image and to de-emphasize (damp) noise. One set of the nearly optimal linear filters are modified second order difference filters in which the number of internal zeros are modified to tune the filter to the particular equipment.