1. A method for determining characteristics of a sample container (1; 31; 32) in an in vitro diagnostic system, the method comprising: in an analysis device (7) of the in-vitro diagnostic system having one or more processors,

providing first image data representing a first image (30) of a sample container (1; 31; 32) in a first scene, wherein in the first scene first illumination conditions comprising background illumination of the sample container (1; 31; 32) are applied to the sample container (1; 31; 32);

providing second image data representing a second image (34) of the specimen container (1; 31; 32) in a second scene, wherein in the second scene a second illumination condition different from the first illumination condition is applied to the specimen container (1; 31; 32);

determining a mask (33) from the first image (30), the mask (33) being indicative of a sub-image section of the first image (30), the sub-image section comprising a representation of the sample container (1; 31; 32) in the first image (30);

determining sub-image data from the second image (34) by applying the mask (33) to the second image (34), the sub-image data comprising a representation of the sample container (1; 31; 32) in the second image (34); and

the characteristics of the sample container (1; 31; 32) are determined from an image data analysis comprising an image data analysis of the sub-image data.

2. The method according to claim 1, further comprising generating third image data representing a third image of the sample container (1; 31; 32), the third image data comprising the sub-image data.

3. The method according to claim 1, further comprising providing the second image data with a second illumination condition of the sample container (1; 31; 32), the second illumination condition being selected from the group of:

in the absence of the background illumination, the ambient illumination,

non-background illumination different from the background illumination alone, such as at least one of top illumination and front illumination, an

A combination of the background illumination and the non-background illumination.

4. The method of claim 3, further comprising applying diffuse illumination to at least one of the background illumination and the non-background illumination.

5. The method according to claim 1, further comprising providing background image data representing a background image (36) of a background scene in which the sample container (1; 31; 32) is absent, wherein the first illumination condition is applied.

6. The method of claim 5, wherein determining the mask (33) comprises processing the first image data and the background image data.

7. The method of claim 6, wherein the processing of the first image data and the background image data comprises comparing a brightness of an image region in the first image (30) with a corresponding image region in the background image (36).

8. The method according to claim 1, further comprising providing reference image data representing a reference image (50) of a background scene missing the sample container (1; 31; 32), wherein the second illumination condition is applied.

9. Method according to claim 8, wherein by means of the reference image (50) a corrected temporary image (51) is generated replacing the second image (34), the generation comprising at least one of:

correcting for non-uniform spatial illumination of the second image (34), an

The second image (34) is spectrally corrected.

10. The method of claim 1, wherein determining the characteristic comprises: for the sample container (1; 31; 32) or a sub-component of the sample container (1; 31; 32), determining at least one of: a geometric property; a color characteristic; a container type; the position of the sample container (1; 31; 32); the state of the sample container (1; 31; 32); the presence or absence of the sample container (1; 31; 32); the presence or absence of a sample in the sample container (1; 31; 32); and information provided on the sample container (1; 31; 32).

11. An analysis device (7) for determining a characteristic of a sample container (1; 31; 32), the analysis device comprising one or more processors configured to:

providing first image data representing a first image (30) of a sample container (1; 31; 32) in a first scene, wherein in the first scene first illumination conditions comprising background illumination of the sample container (1; 31; 32) are applied to the sample container (1; 31; 32);

providing second image data representing a second image (34) of the specimen container (1; 31; 32) in a second scene, wherein in the second scene a second illumination condition different from the first illumination condition is applied to the specimen container (1; 31; 32);

determining a mask (33) from the first image (30), the mask (33) being indicative of a sub-image section of the first image (30), the sub-image section comprising a representation of the sample container (1; 31; 32) in the first image (30);

determining sub-image data from the second image (34) by applying the mask (33) to the second image (34), the sub-image data comprising a representation of the sample container (1; 31; 32) in the second image (34); and

the characteristics of the sample container (1; 31; 32) are determined from an image data analysis comprising an image data analysis of the sub-image data.

12. An in vitro diagnostic system comprising an analysis device (7) with one or more processors, the analysis device being configured to:

providing first image data representing a first image (30) of a sample container (1; 31; 32) in a first scene, wherein in the first scene first illumination conditions comprising background illumination of the sample container (1; 31; 32) are applied to the sample container (1; 31; 32);

providing second image data representing a second image (34) of the specimen container (1; 31; 32) in a second scene, wherein in the second scene a second illumination condition different from the first illumination condition is applied to the specimen container (1; 31; 32);

determining a mask (33) from the first image (30), the mask (33) being indicative of a sub-image section of the first image (30), the sub-image section comprising a representation of the sample container (1; 31; 32) in the first image (30);

determining sub-image data from the second image (35) by applying the mask (33) to the second image (34), the sub-image data comprising a representation of the sample container (1; 31; 32) in the second image (35); and

the characteristics of the sample container (1; 31; 32) are determined from an image data analysis comprising an image data analysis of the sub-image data.

13. An in vitro pre-analysis system comprising an analysis device (7) with one or more processors, the analysis device being configured to:

providing first image data representing a first image (30) of a sample container (1; 31; 32) in a first scene, wherein in the first scene first illumination conditions comprising background illumination of the sample container (1; 31; 32) are applied to the sample container (1; 31; 32);

providing second image data representing a second image (34) of the specimen container (1; 31; 32) in a second scene, wherein in the second scene a second illumination condition different from the first illumination condition is applied to the specimen container (1; 31; 32);

determining a mask (33) from the first image (30), the mask (33) being indicative of a sub-image section of the first image (30), the sub-image section comprising a representation of the sample container (1; 31; 32) in the first image (30);

determining sub-image data from the second image (34) by applying the mask (33) to the second image (34), the sub-image data comprising a representation of the sample container (1; 31; 32) in the second image (34); and

the characteristics of the sample container (1; 31; 32) are determined from an image data analysis comprising an image data analysis of the sub-image data.

Background

The in vitro diagnostic system is suitable for testing samples, such as blood or tissue samples taken from the human body. In vitro diagnostics can detect diseases or other disorders and can be used to monitor the overall health of a person to help cure, treat, or prevent a disease. In vitro diagnostics may also be applied in sophisticated medicine to determine which patients may benefit from a particular treatment or therapy. Some in vitro diagnostic tests are used in laboratories or other health professional institutions.

Sample containers (also referred to as sample vessels), such as blood collection tubes, and other containers, such as urine, swabs, tissue, bone marrow, capillaries, etc., are now commonly handled and handled by automated solutions (in vitro diagnostic systems). Such automated solutions require, for reliable operation, a sample container analysis apparatus (SCAM) to provide at least one of the following: identifying the presence of a sample vessel; identifying the state of the sample vessel, such as open and closed, up and inverted, labeled and unlabeled, rotated and not rotated; reporting the position and/or pose of the container; reporting the position and/or attitude of a part associated with the container, such as the position and/or attitude of a label, lid, or orientation mark; identifying a type (TTI) of a sample container; and identifying irregularities of or on the sample container. This is a clear object localization of the sample container or vessel in the captured image, allowing for a clear propagation of the outline of the sample container, e.g. in order to derive the exact geometry, state, etc.

Current sample vessel analysis devices (SCAMs) are in many cases difficult to operate reliably. This situation may lead to unexpected downtime, requiring manual intervention steps to remedy the shortcomings of these systems.

Document US 2019/0033209 a1 discloses a model-based method for quantifying a specimen. The method comprises the following steps: providing a sample; capturing images of the sample under multiple spectral illuminations of different nominal wavelengths and exposures; and classifying the sample into various category types including one or more of serum or plasma fractions, precipitated blood fractions, gel separators (if used), air, tubing, labels, or bottle caps; and quantifying the sample. The quantifying includes determining one or more of: a location of a liquid-air interface, a location of a serum-blood interface, a location of a serum-gel interface, a location of a blood-gel interface, a volume and/or depth of a serum or plasma fraction, or a volume and/or depth of a precipitated blood fraction. Quality inspection modules and sample testing apparatus adapted to perform the method are described, as well as other aspects. Furthermore, document US 2018/0365530 a1 mentions a model-based method of determining characteristics of a sample container.

Document US 2018/0364268 a1 discloses a model-based method of classifying samples in sample containers.

Document WO 2019/018314 a1 discloses an apparatus for characterizing samples and/or sample containers. The characterization apparatus includes: an imaging location configured to receive a specimen container containing a specimen; a light source configured to provide illumination of an imaging location; and a hyperspectral image capture device. The hyperspectral image capture device is configured to generate and capture at the spectral image capture device a spectrally resolved image of the specimen container and a small portion of the specimen. Processing, by a computer, spectrally resolved image data received at a spectral image capture device to determine at least one of: partitioning of at least one of the samples and/or sample containers; and determination of the presence or absence of an interferent (such as hemolysis, jaundice, or lipemia). Methods of imaging a specimen and/or specimen container and a specimen testing apparatus including a characterization apparatus, among other aspects, are described.

WO 2018/022280 a1 discloses a model-based method of determining characteristics of a specimen container bottle cap to identify the container bottle cap. The method comprises the following steps: providing a specimen container including a container cap; capturing backlight images of container closures taken at different exposure lengths and using a plurality of different nominal wavelengths; selecting pixels of optimal exposure from the images of different exposure lengths for each nominal wavelength to generate image data of optimal exposure for each nominal wavelength; classifying the best exposure pixels as at least one of a tube, a label, or a bottle cap; and identifying a shape of the container closure based on the image data for each nominal wavelength and the best exposure pixels classified as closures. A quality inspection module and specimen testing apparatus adapted to perform the method are described, along with a number of other aspects.

ZaydeEt al describe image segmentation (separation of object from background) in "development and application of dual image method for accurate object recognition and color analysis" ("journal of food engineering, epskshire, bankin, uk, volume 111, phase 1, month 1, 28 days 2012-01-28, pages 46-51). Specifically, for objects similar to the background color, the "two-image" method is applied. The method comprises the following steps: defining an object from silhouettes obtained from the backlight image; and performing color analysis using the pre-image segmented from the backlight image.

Document WO 2018/089935 a1 refers to a characterization device comprising pattern generation. In some embodiments, the characterization apparatus is configured to characterize the sample and/or sample container.

US 7840360B 1 discloses a system and method for non-invasive inspection of one or more infrared light transmissive vessels containing a liquid using a Near Infrared (NIR) imaging device in combination with one or two NIR light sources, a diffuser plate and an optical wavelength selection device provided for selecting one or more wavelength bands.

Disclosure of Invention

It is an object of the present disclosure to provide a method, an analysis device and an in-vitro diagnostic system for determining characteristics of a sample vessel (sample container), which support a safe and efficient handling of the sample vessel in the in-vitro diagnostic system.

To solve this problem, a method for determining characteristics of a sample vessel in an in vitro diagnostic system according to claim 1 is provided. Furthermore, an analysis device and an in vitro diagnostic system according to claims 11 and 12, respectively, are provided. In addition, an in vitro pre-analysis system according to claim 13 is provided. Further embodiments are disclosed in the dependent claims.

According to one aspect, there is provided a method for determining characteristics of a sample vessel (sample container) in an in vitro diagnostic system, the method comprising the steps of: in an analysis apparatus of an in vitro diagnostic system having one or more processors, providing first image data representing a first image of a sample vessel in a first scene, wherein in the first scene a first illumination condition comprising background illumination of the sample vessel is applied to the sample vessel; providing second image data representing a second image of the sample vessel in a second scene, wherein in the second scene a second illumination condition different from the first illumination condition is applied to the sample vessel; determining a mask or mask image from the first image, the mask indicating a sub-image section of the first image, the sub-image section comprising a representation of the sample vessel in the first image; determining from the second image, by applying a mask to the second image, a sub-image data/image data subset containing a representation of the sample vessel in the second image; and determining a characteristic of the sample vessel from an image data analysis, the image data analysis including an image data analysis of the sub-image data.

According to another aspect, there is provided an analysis apparatus for determining characteristics of a sample vessel (sample container), the analysis apparatus comprising one or more processors configured to: providing first image data representing a first image of a sample vessel in a first scene, wherein in the first scene a first illumination condition comprising background illumination of the sample vessel is applied to the sample vessel; providing second image data representing a second image of the sample vessel in a second scene, wherein in the second scene a second illumination condition different from the first illumination condition is applied to the sample vessel; determining a mask from the first image, the mask indicating a sub-image section of the first image, the sub-image section comprising a representation of the sample vessel in the first image; determining sub-image data from the second image by applying a mask to the second image, the sub-image data comprising a representation of the sample vessel in the second image; and determining a characteristic of the sample vessel from an image data analysis, the image data analysis including an image data analysis of the sub-image data.

According to yet another aspect, there is provided an in vitro diagnostic system comprising an analysis device having one or more processors, the analysis device configured to: providing first image data representing a first image of a sample vessel in a first scene, wherein in the first scene a first illumination condition comprising background illumination of the sample vessel is applied to the sample vessel; providing second image data representing a second image of the sample vessel in a second scene, wherein in the second scene a second illumination condition different from the first illumination condition is applied to the sample vessel; determining a mask from the first image, the mask indicating a sub-image section of the first image, the sub-image section comprising a representation of the sample vessel in the first image; determining sub-image data from the second image by applying a mask to the second image, the sub-image data comprising a representation of the sample vessel in the second image; and determining a characteristic of the sample vessel from an image data analysis, the image data analysis including an image data analysis of the sub-image data.

According to yet another aspect, an in vitro pre-analysis system is provided, comprising an analysis device having one or more processors, the analysis device being configured to provide first image data representing a first image of a sample vessel in a first scene, wherein in the first scene a first illumination condition comprising background illumination of the sample vessel is applied to the sample vessel; providing second image data representing a second image of the sample vessel in a second scene, wherein in the second scene a second illumination condition different from the first illumination condition is applied to the sample vessel; determining a mask from the first image, the mask indicating a sub-image section of the first image, the sub-image section comprising a representation of the sample vessel in the first image; determining sub-image data from the second image by applying a mask to the second image, the sub-image data comprising a representation of the sample vessel in the second image; and determining a characteristic of the sample vessel from an image data analysis, the image data analysis including an image data analysis of the sub-image data.

The proposed technique provides a more reliable identification of at least one of sample vessels and properties and poses in digital images of sample vessels taken for different scenes. Based on the first image, a mask of the sample vessel is determined from the first image. The mask of the vessel indicates the contour line of the sample vessel, which may also be referred to as the contour of the sample vessel in some embodiments. Subsequently, the sample characteristics can be determined with improved accuracy. Determining the properties with higher reliability will support safer operation of the in-vitro diagnostic system.

In alternative embodiments, the characteristics of more than one sample vessel(s) may be determined. A common characteristic of all sample vessels can be determined. Alternatively, different characteristics of different sample vessels may be determined.

The method may further include generating third image data representing a third image of the sample vessel, the third image data including sub-image data. The third image will display at least image data from the image area defined by the application mask (outline or contour of the sample vessel). In some embodiments, it may be allowed to independently process image data of the third image for determining the characteristic in the digital image analysis.

The second image data may be provided with a second illumination condition selected from the group of: no background illumination; non-background lighting other than only background lighting, such as at least one of top lighting or front lighting; and combinations of background and non-background illumination. One or more of the lighting conditions from the set may be applied.

The method may further include applying diffuse illumination under at least one of background illumination and non-background illumination.

Background image data may be provided, the background image data representing a background image of a background scene lacking the sample vessel, wherein in the background scene the first illumination condition is applied. The background scene may correspond to a first scene where the first illumination condition is applied, but the sample vessel is absent or absent.

In particular, the mask may indicate which part of the first image is related to the sample vessel and which part of the first image is not related to the sample vessel.

Determining the mask may include processing the first image data and the background image data.

The processing of the first image data and the background image data may comprise comparing the brightness of an image region in the first image with a corresponding image region in the background image.

The determining of the mask may comprise determining a transmission value for each point/pixel in the first image. The determining may include comparing the first image to a background image. For example, the first image and the background image may be compared by comparing the images pixel by pixel. It may be provided that an image pixel in the mask is assigned to an image area associated with the sample vessel if the transmission value (of the image pixel) is less than a threshold value, for example about 90%, about 95% or about 98%, otherwise the pixel is assigned to an image area (background) not associated with the sample vessel. By determining the transmission value, it is also possible to detect a transparent body or section of the sample vessel.

Reference image data may be provided, the reference image data representing a reference image of a background scene lacking the sample vessel, wherein the second illumination condition is applied. The reference object may be provided at a position where the sample vessel is located in the first scene and/or the second scene. The reference object may be a flat material object, preferably having a defined reflectivity. Alternatively, the surface of the background illumination (device) can be applied as a reference object.

Provision can be made for a corrected temporary image to be generated which replaces the second image by means of the reference image. The generating may include correcting for non-uniform spatial illumination of the second image and/or spectrally correcting the second image.

The determination of the characteristic may comprise, for the sample vessel or a sub-portion of the sample vessel, determining at least one of: a geometric property; a color characteristic; a vessel type; the posture of the sample vessel; the presence or absence of a sample in the sample vessel; the status of the sample in the sample vessel; a sampling state of the sample in the sample vessel; and information provided on the sample vessel.

The embodiments disclosed with respect to the method for determining the characteristics of the above-described sample vessel may be applied to at least one of an analysis device, an in-vitro diagnostic system and an in-vitro pre-analysis system, with suitable modifications in detail.

Drawings

Embodiments are described below, by way of example, with reference to the accompanying drawings. Shown in the drawings are:

fig. 1 is a schematic representation of an arrangement for detecting images of sample vessels provided in different scenes applying different illumination conditions for the sample vessels;

fig. 2 is a schematic representation of another arrangement for detecting images of sample vessels provided in different scenes applying different illumination conditions for the sample vessels;

FIG. 3 is a schematic representation of different images provided in a method for determining characteristics of a sample vessel in an in vitro diagnostic system;

fig. 4 another schematic representation of different images provided in a method for determining characteristics of sample vessels in an in vitro diagnostic system, wherein images of a plurality of sample vessels are taken; and

fig. 5 is a schematic representation of different images provided in a method for determining characteristics of a sample vessel in an in vitro diagnostic system using flat field and/or white balance correction.

Detailed Description

Fig. 1 and 2 show schematic representations of an arrangement for detecting and processing (digital) images of sample vessels 1 provided in different scenes in which different illumination conditions are applied to the sample vessels 1. The sample vessel 1 may receive (test) a sample, such as a blood or tissue sample taken from a human body. Diseases or other conditions may be detected by analyzing the test sample received in the sample vessel 1 in the in vitro diagnostic system.

The arrangement depicted in fig. 1 and 2 may be used in an analysis device provided in an in vitro diagnostic system to detect a plurality of images of the sample vessel 1 in different scenes which are distinguished at least by different illumination conditions applied to the sample vessel 1.

After detection, the images detected by the arrangement in fig. 1 and 2 may be analyzed by image data analysis or processing to determine characteristics of the sample vessel 1 (i.e. characteristics of the sample vessel). The information about the characteristic may be processed or used in an in vitro diagnostic system to operate the system. Thus, a reliable determination of the properties will enable a safe handling of the sample vessel 1 and the sample received in the sample vessel 1 in an in vitro diagnostic system. For example, in response to determining a specific characteristic, some special program of sample analysis may be provided in the in vitro diagnostic system with the sample vessel 1, which special program is assigned to the determined characteristic of the sample vessel 1.

For the arrangement depicted in fig. 1 and 2, there is a background illumination device 2 for providing background illumination for the sample vessel 1. The background illumination means 2 mainly directs light to the back of the sample vessel 1 or to the part of the sample vessel 1 being analyzed. Further, for example, lighting devices 3a, 3b for applying diffuse illumination are provided. The background illumination means 2 may also be provided with a light source for diffuse background illumination. In an alternative embodiment, at least one of the background illumination means 2 and the illumination means 3a, 3b may be configured to provide non-diffuse illumination of the sample vessel 1. The illumination means 3a, 3b may be used for front and/or top illumination of the sample vessel 1, providing the option of applying different illumination conditions to the sample vessel 1 in different scenes.

In addition, in the embodiment shown, the background illumination means 2 extends over the entire size of the sample vessel 1. For example, the background illumination device 2 may be provided with: self-luminous lighting panels, for example, lighting panels based on LEDs (LED-light emitting diodes) or OLEDs (LED-organic light emitting diodes); or a non-self-luminous surface, which is indirectly illuminated. A similar configuration may be provided for the lighting devices 3a, 3 b. The wavelength of the illumination is selected according to the application. Simple presence detection may use narrow bandwidth light, such as light from a colored LED. White light may be required for color determination of the bottle cap.

A detection device 4 is provided for detecting or acquiring (digital) images of the sample vessel 1 when the sample vessel 1 is illuminated by the background illumination device 2 and/or the illumination devices 3a, 3 b. In addition, in the case where the light sources of the background illumination device 2 and the illumination devices 3a, 3b do not illuminate the sample vessel 1, the image can be detected by the detection device 4. The camera provided with the detection means 4 may be a "grayscale" camera, an RGB camera, or may be a camera with a specific sensitivity, depending on the purpose or application. The detection means 4 may also have time-of-flight capabilities or may be a stereo camera in order to retrieve also 3D data or distance information. The back or background illumination means 2 is placed at a greater distance than the sample vessel or container 1 with respect to the detection means 4.

The background illumination means 2, the illumination means 3a, 3b and the detection means 4, which may be provided with a digital camera, are connected to a control means 5 configured to control the operation of the different means while detecting images of the sample vessel 1 for different scenes. During image acquisition, the control means 5 control the illumination of the scene and the detection means 4 by at least one of the background illumination means 2 and the illumination means 3a, 3 b. This may include any of component triggering, powering, synchronization, and parameterization. The control device 5 may also be configured to process images according to different embodiments disclosed herein.

For the different arrangements in fig. 1 and 2, there are different types of optical elements 6, such as field lenses (see fig. 1) and cylindrical lenses (see fig. 2), for providing optimized optical conditions depending on the scene to be detected.

An analyzing means 7, which may be implemented together with the control means 5 and comprises one or more processors for processing (digital) image data, is connected to the detection means 4. The plurality of images detected by the detection means 4 will be processed in the analysis means 7 to determine the characteristics of the sample vessel 1. The analysis device 7 may be connected to a virtual host or server device (not shown) for data communication and synchronization.

Fig. 3 and 4 each show a schematic representation of a plurality of (digital) images provided in a method for determining a property of one or more sample vessels from an image detected, for example, by one of the arrangements shown in fig. 1 and 2. The plurality of images detected by the detection means 4 are processed by the analysis means 7, as described below for the embodiments. According to the embodiment in fig. 3, in a first image 30, a (single) sample vessel 31 is shown, representing a single sample vessel such as the sample vessel 1 depicted in fig. 1 and 2.

Fig. 4 shows another schematic representation of different images provided in a method for determining characteristics of sample vessels in an in vitro diagnostic system, wherein images of a plurality of sample vessels 32 are provided. According to fig. 4, in an alternative embodiment, a plurality of sample vessels 32 is shown in the first image 30, the plurality of sample vessels 32 representing a plurality of sample vessels, such as a plurality of sample vessels 1, each for receiving a sample. Fig. 4 also shows how the uniform transparent section of the sample vessel 32 is well shown by applying front illumination.

The first image 30 represents an image of the sample vessel 31 in a first scene in which background illumination is applied to the sample vessel 31. Similarly, the first image 30 in fig. 4 represents an image of a plurality of sample vessels 32 in a first scene with background illumination provided by the background illumination device 2 applied.

A mask 33 for the sample vessel 31/sample vessels 32 is provided, the mask 33 representing the outline or outline of the sample vessel 31 or sample vessels 32. The mask 33 is generated by image data analysis or processing of the first image 30. The mask 33 comprises an area 33a related to the sample vessel 31, 32 and an area 33b unrelated to the sample vessel 31, 32 but related to the background.

Still referring to fig. 3 and 4, a second image 34 is shown showing the sample vessel 31/plurality of sample vessels 32 in a second scene to which a second illumination condition is applied, the second illumination condition being different from the first illumination condition used for detecting the first image 30. In the embodiment shown in fig. 4, the second lighting condition applies both the background lighting provided by the background lighting device 2 and (in addition to the first scene) the top lighting and/or the front lighting provided by the lighting devices 3a, 3 b.

Mask 33 is applied to second image 34 to process mask image 35, which includes image data from a subset of the image data in second image 34, where the subset of image data is defined by mask 33. As a result of the described procedure, the (tube) image of the sample vessel 31/sample vessels 32 depicted by the image data subset is clearly separated from the background, is full-color, and is free of reflections and shadows. Thus, in particular, the mask image 35 enables simple and reliable image analysis, such as for vessel pose and vessel type recognition.

Subsequently, the mask image 35 may be processed to determine characteristics of one or both of the sample vessels 31, 32. For example, the characteristics to be determined by image data analysis may include one or more of the following characteristics: the presence or absence of a sample vessel; the location of the sample container; the posture of the sample vessel; a pinch point of the sample vessel; the size of the sample container; the status of the sample container, e.g., filled or unfilled; properties of the container part, such as the colour of the cap, the length of the tube, the level; and a sample container type. Such data or image analysis of the mask image 35 is known as such.

The process or method outlined above provides a more reliable determination of the characteristic from the mask image 35 derived from the image data processing based on the first image 30, the second image 34 and the mask 33.

According to the embodiment shown in fig. 3 and 4, the background image 36 is detected by the detection means 4 of the background scene. For a background scene, a first illumination condition is applied, wherein background illumination is provided by the background illumination means 2. In the background scenario, any sample vessels were missing or absent.

The background image 36 and the first image 30 are processed to determine the mask 33. The determination of the mask 33 may include determining a transmission value for each point, each pixel, or each segment in the first image 30. The first image 30 is compared to the background image 36. For example, the first image 30 and the background image 36 may be compared by comparing the images on a pixel-by-pixel basis. It may be provided that if the transmission value (of an image pixel) is less than a threshold value, for example about 90%, about 95% or about 98%, the corresponding image pixel in the mask 33 is assigned to an image region associated with the sample vessel 31, 32, otherwise the pixel is assigned to an image region (background) not associated with the sample vessel 31, 32. By determining the transmission value, the transparent body can also be detected.

For example, at least one of the first image 30, the mask 33, the second image 34, the background image 36, and the reference image 50 may be stored in a memory of the control device 5 or a memory accessible to the control device 5.

Fig. 5 shows a schematic representation of different images provided in a method for determining characteristics of a sample vessel in an in vitro diagnostic system using flat field and/or white balance correction.

In the background image 36, a triangular symbol 37 is depicted, which schematically represents a real or virtual object in the background lighting scene, e.g. lighting non-uniformity. Since a real or virtual object (symbolically represented by a triangle 37) is present both in the background image 36 and in the first image 30 showing the sample vessel 31, it does not negatively influence the mask 33.

In the second image 34, a circle symbol 38 is shown, which schematically represents, for example, illumination inhomogeneities, in particular in the case of application of front lighting of the lighting devices 3a, 3 b. Such illumination non-uniformity may refer to, for example, lower brightness in the edge region. This may be due to the applied illumination means 3a, 3b and/or imaging optics.

A reference image 50, which may also be referred to as a second background image, is provided for compensating for the non-uniformities schematically represented by the circle symbol 38. The reference image 50 and the second image 34 are processed to determine a corrected temporary image 51 corrected for non-uniformities schematically represented by the circle symbol 38. Next, a mask image 35 is determined by processing the mask 33 and the corrected temporary image 51.

Regarding the unevenness indicated by the circle symbol 38, a white plate may be used as a reference, which may reflect the difference in exposure "beautifully". Since the (illumination) non-uniformity varies over time (as long as the settings do not vary), it is sufficient to capture and store the reference image 50 once. The brightness distribution of the reference image 50 without the sample vessel 31 is then used as a compensation or correction. For example, if the brightness of an image subregion is only half of another image subregion than the image subregion, the intensity (pixel) values of the image subregion are scaled up accordingly. Furthermore, if the neutral white image sub-region does not show a nominal R: G: B ratio (e.g., 1:1:1), these R: G: B ratios may be normalized (which technically corresponds to white balance). The latter is used to correct color errors caused by illumination or color reproduction of the detection device 4 such as a digital camera.

List of reference numerals

1 sample vessel

2 background lighting device

3a, 3b lighting device

4 detection device

5 control device

6 optical element

7 analysis device

30 first image

31 sample vessel (shown in the figure)

32 multiple sample vessels (shown in the figure)

33 mask

33a area relating to a sample vessel

33b area independent of the sample vessel

34 second image

35 mask image

36 background image

37 triangle symbol

38 circle symbol

50 reference image

51 corrected temporary image