GB2537887A - System and method for identifying a sample container, a reagent container or a rack - Google Patents

Abstract

A system for identifying a sample container or a reagent container or a rack 140, the system comprising loading module 110, a camera 170, an LED bar 120 comprising at least one LED, at least one mirror 160, at least one control PCBA and a sledge 190 wherein the camera or the at least one mirror is mounted on the sledge with the at least one mirror positioned between the camera and the container or rack so that the optical path of the camera is reflected at least once between the camera and the container or the rack to enable the camera to record or film at least one picture or frame of the container or rack as it is inserted into the module. The sledge is motor-driven and moves to position the camera or the at least one mirror. The LED bar is positioned to illuminate the container. Ideally, a motion or proximity sensor detects a users presence near the loading module so that the system may be activated. A method of operation is also disclosed.

Description

Title: -1-

Description

System and method for identifying a sample container, a reagent container or a rack

Field of the invention

[0001] The present invention is directed to a system and a method for identifying a sample container, a reagent container or a rack.

Background of the invention

[0002] Automated analyzer systems typically incorporate modules for loading of e.g. patient samples and assay-specific reagents in reagent containers Loading is mostly done manually, meaning that the speed with which a target like a code is passing a means of identification like a scanner can widely vary. Typically, these modules are equipped with laser scanners that allow for bar code reading. Bar codes are the most widely used means of identification for patient samples at a doctor's practice, in laboratories, hospitals, clinical applications and blood banks.

[0003] Laser scanners are commonly used for bar code scanning. They differ in a wide range of technical specifications and price. Reading speeds and the ability to focus at various distances or only providing a fix focus shall be mentioned as technical specifications exemplarily.

[0004] 2D codes are becoming more and more popular in this context as they allow for a denser packaging of information in a smaller area than bar codes. 2D codes are already common for reagent IDs and there is an upcoming demand to have 2D codes for patient sample identification on analyzer systems as well.

[0005] 2D code readers are typically cameras with a CCD or CMOS chip. They also differ widely in technical aspects like resolution, detection range and reading speeds, the ability to focus, integrated illumination, etc [0006] Beyond the only identification of sample containers or reagent containers on racks it is beneficial to identify their location or their exact position and the order of sample containers or reagent containers on racks. In addition, there are many other features and specifications available on racks comprising sample containers and/or reagent containers

depending on instrument specifications and needs.

[0007] A module for sample container or reagent container loading and identification within an analyzer instrument typically provides an interface to the user as well as to the system. Disadvantages of common systems for identification are: Slow laser line scanners and 2D scanners can only handle non-or slowly moving objects. Users are accustomed to faster loading speeds when introducing a new rack. However, fast laser-based bar code scanners are not feasible.

IS In case of manual loading and separated identification (identification is taking place after manual insertion), the user does not get an initial feedback on his loading procedure, e. g. if every sample container/reagent container has been identified successfully. If additional action is needed for a correct identification, e. g. reloading of a rack or repositioning of tubes or correct labeling, feedback to the user will be delayed.

As a result, manual loading gets annoying and more time-consuming for the user, The instrument design is driven towards higher automation effort, bigger module envelopes and higher effort and cost -Most or practically all scanners are not able to read within a deep focus range without losing resolution and/or luminosity. As a result, in a large loading bay with a high number of rack lanes, either the scanner has to be moved to reach all targets (i.e. the sample container, the reagent container or the rack), or the targets have to move towards the scanner, or the targets further away from the scanner can only be read with reduced resolution.

Most or practically all scanners need a defined minimum distance to their closest target to achieve a minimum reading height / width for identification In most cases, the specified reading height in analyzer application is driven by the CLSI standard AUT02-A2 Off-the-shelf components are limited in terms of what codes or features they can identify, depending on design and technology used and on possibilities for (e. g. firmware-) upgrades. Most of the available components can only detect bar codes, some additionally 2D codes.

Some scanners provide an integrated target (code) illumination. This illumination is sometimes helpful but typically critical in terms of long distance reads and total reflection Total reflection is highly depending on shape and reflectivity of scanned surfaces and arises easily with illumination means positioned close to the camera lens. Any total reflection creates data loss on 2D codes.

Solutions working with polarized light for target illumination can reduce the risk of total reflection on the target surface but lead to significant light losses. A factor of 3 to 4 of increased light intensity is needed to compensate for these losses and leads to bulky installations and higher cost.

[0008] There is currently no off-the-shelf solution available for fast feature identification and bar code and 2D code detection with a homogeneous and ready-to-install reliable illumination over a wide distance range that can create instant decoding results for feedback to the instrument and the user.

Object of the Invention [0009] It is thus an object of the present invention to provide a system and a method for identifying a sample container or a reagent container or a rack, which can handle fast-moving insertion of the sample container, the reagent container or the rack and where the user gets initial feedback on the loading, i.e, the insertion procedure. Identification shall be possible within a deep focus range without losing resolution and/or luminosity and without the problem of light reflection creating data loss.

Summary of the Invention

[0010] A system for identifying a sample container or a reagent container or a rack is provided. The system comprises a loading module, a camera, an LED bar comprising at least one LED, at least one mirror, at least one control PCBA and a sledge, wherein the camera or the at least one mirror is mounted on the sledge, wherein the at least one mirror is positioned between the camera and the at least one of the sample container, the reagent container and the rack so that the optical path of the camera is reflected at least once between the camera and the at least one of the sample container, the reagent container and the rack.

[0011] The rack may comprise at least one of a sample container and a reagent container.

[0012] In one aspect of the invention, the at least one mirror is a fully reflective mirror or a semi-transparent mirror.

[0013] The system may further comprise a sensor for sensing an approximation of a user to the system [0014] The sensor may be a motion sensor or a proximity sensor.

[0015] In one aspect of the invention, the loading module comprises at least one insertion lane for at least one rack.

[0016] The sledge may be motor-driven.

[0017] In another aspect of the invention the system further comprises a data processing unit and an image processing software [0018] The camera comprised in the system may comprise a CCD or a CMOS chip.

[0019] The lens of the camera may have a fixed focal length and a fixed focus.

[0020] A width of the loading module may be up to 300 mm when the focal length of the lens is 25 mm and the width of the loading module may be up to 500 mm when the focal length of the lens is 35 mm.

[0021] In one aspect of the invention, the PCBA comprises at least one detector and controls at least one of the LED bar, the camera and the sledge.

[0022] A method for identifying a sample container or a reagent container or a rack is provided The method comprises the following steps focusing the camera of the system on the sample container or the reagent container or the rack being inserted into or retracted from the loading module of the system by moving the camera or the at least one mirror of the system on the sledge of the system and directing the optical path of the camera to the at least one of the sample container, the reagent container and the rack, illuminating the sample container or the reagent container or the rack by the LED bar of the system, recording at least one picture by the camera, wherein the picture comprises a frame covering at least part of the sample container or the reagent container or the rack.

[0023] The method may further comprise as a first step sensing an approximation of a user by the sensor and thereby starting the sequence of method steps [0024] In one aspect of the invention, a code is located on at least one of the sample container, the reagent container and the rack for identifying the sample container or the reagent container or the rack.

[0025] The at least one LED of the LED bar may be pulsed at a multiple of a frame rate of the camera.

Summary of the Figures

[0026] The invention will now be described on the basis of the drawings. Tt will be understood that the embodiments and aspects of the invention described herein are only examples and do not limit the protective scope of the claims in any way. The invention is defined by the claims and their equivalents. It will be understood that features of one aspect or embodiment of the invention can be combined with a feature of a different aspect or aspects and/or embodiments of the invention. It shows: [0027] Figure 1 Exemplary system according to the present invention.

[0028] Figure 2 System according to the present invention comprising a sensor.

[0029] Figure 3 System with a moving camera.

[0030] Figure 4 System with a moving reflective mirror.

[0031] Figure 5 System with a moving reflective mirror and a semi-transparent mirror.

[0032] Figure 6 Field of view on target

[0033] Figure 7 Possible illumination concept of the present invention.

Detailed Description of the Invention and the Figures [0034] The present invention provides a system for identifying a sample container or a reagent container or a rack, wherein the system comprises a loading module, a camera, an LED bar comprising at least one LED, at least one mirror, at least one control PCBA and a sledge, wherein the camera or the at least one mirror is mounted on the sledge, wherein the at least one mirror is positioned between the camera and the at least one of the sample container, the reagent container and the rack so that the optical path of the camera is reflected at least once between the camera and the at least one of the sample container, the reagent container and the rack [0035] The system may be an automated analyser system, where e. g. patient sample containers and reagent containers can be inserted within racks. In sub-modules of the system, the sample containers, reagent containers and racks and potentially other consumables may be loaded and identified. Said sub-modules provide means for reading or scanning information that is located on the sample containers or reagent containers like sample Ins, types of reagents, types of consumables, expiration dates, number of tests to be performed out of carriers etc. The sub-modules may be comprised in the system but they may also be a further component added to the system.

[0036] Within the meaning of the present invention, the sample container, the reagent container and the rack are also referred to as the -target" because these items are to be scanned and identified. The camera is the means of identification and scans and identifies the targets. The terms "reading" and "scanning" shall be used synonymously.

[0037] The term "sample container" refers to any kind of liquid, solid or gaseous matter in any kind of container that is to be tested in an analysing system. The sample container may also comprise, for example, body fluid from a patient or from a healthy individual [0038] The term "reagent container" refers to any kind of container which contains a reagent. The reagent may be necessary for any kind of testing in the analysing system, it may for example be a chemical agent.

[0039] Containers may be capped or uncapped. Containers may comprise glass, plastic tubes or bottles.

[0040] The rack may be empty or may comprise at least one sample container and/or at least one reagent container.

[0041] Thus, the invention describes a solution to allow for reading of all typically used bar codes plus all 2D code types (by way of example only: QR-codes, Matrix-codes) with a camera-based system. The solution supports detection with moving target speeds comparable to those of the most expensive currently implemented laser bar code scanners at a wider scanner range (closest to furthest distance from scanner to target) without any loss of resolution, quality or luminosity over the full reading distance.

[0042] The loading module may comprise at least one insertion lane for at least one rack.

The illumination concept of the present invention comprises the activation of the LED bar. The illumination concept is scalable to allow for small and large loading modules with a respective low and high number of rack insertion lanes and consistent lighting of the targets. Loading modules with one rack insertion lane up to 12 and more rack insertion lanes are possible.

[0043] The sledge may be motor-driven so that the camera or the at least one mirror mounted on the sledge can easily be moved back and forth to adjust the camera or the at least one mirror to the target.

[0044] Figure 1 shows an exemplary system according to the present invention which comprises a loading module 110, an LED bar 120 for illumination purposes, a rack 140, a camera 170 on a motor-driven camera sledge 190, a sledge drive 180, a tilted mirror 160, protective glass 150, control PCBAs and an image processing software running on the instrument PC. The camera optical path 130 is also shown.

[0045] The hardware is assembled to an exemplary loading module, i.e. to a system according to the present invention. The illustrated system in Figure 1 provides 12 insertion lanes 145 for racks 140 at a lane to lane distance of 25 mm. However, for the described invention the width of the insertion area is scalable and can be narrower and wider. One rack 140 is partially inserted in the rearmost lane in Figure 1.

[0046] Under a cross-bar at the top front of the system an LED bar 120 is installed. Once a rack 140 shall be loaded into the system, a defined number of LEDs (for example, 2 to 3) at one side of the inserted rack 140 is activated. In the illustrated case, the LEDs to the right of the insertion lane 145 of the inserted rack 140 illuminate the code labels on the right side (in insertion direction) of the rack 140, and the camera 170 takes picture frames via a tilted mirror 160 and through a protective glass 150 along its optical path 130 [0047] The lens of the camera 170 in Figure 1 is equipped with a fixed focal length and a fixed focus. However, a fixed focal length and a fixed focus are not necessary in the sense of the present invention The focusing plane of the camera 170 in Figure 1 is set to the desired insertion lane by moving the camera 170 on a motor-driven moveable camera sledge 190 along the side of the loading module 110.

[0048] LED bar 120, camera 170 and camera sledge drive 180 in Figure 1 are controlled by a processor PCBA that is located at the rear of the system. The PCBA may also comprise at least one detector, for example to detect fully inserted racks comprising at least one of a sample container and a reagent container. The camera picture frames are routed through this PCBA to a PC with an image processing software for decoding [0049] Importantly, identification of the target does not only comprise code reading or scanning. The camera may take a picture sequence and the sequence together with the analysing software allow for recognition of various kinds of features. Useful in this context are for example: the determination of the exact position and/or the order of sample containers or reagent containers within racks - the detection of presence or absence of sample containers or reagent containers in racks (detection of empty / unloaded positions). This is beneficial because the software may distinguish between an empty position and one loaded with an unreadable or missing identifier code.

- the possibility for differentiation of types of racks the possibility for differentiation of types of containers (by way of example only: glass, plastic tubes, bottles) loaded within a rack the detection of the size of containers (diameters / heights / widths) the detection of a fill level within a container the chance for a differentiation between capped or uncapped containers the detection of types and sizes of caps for uncapping devices.

[0050] Advantageously, additional codes on racks are not necessary in order to identify, for example, an empty position in a rack or to identify the order of sample containers or reagent containers in racks because the camera can identify many features without using codes.

[0051] Thus, according to the present invention the purpose of 'identifying a sample container or a reagent container or a rack comprising at least one of a sample container and a reagent container" shall comprise identifying at least one of a code, a size and/or type of at least one of a sample container, a reagent container and a rack, a fill level of a container, a size and/or type of a cap of a reagent container, a position of at least one of a sample container and a reagent container and a presence of at least one of a sample container and a reagent container.

[0052] The solution may further comprise an intelligent sensing concept when to initiate the camera's picture sequence to not cause unnecessary computing time because every image that arrives at the image processing software uses up computer time. Sensors may be installed to sense the approximation of a user when introducing a new rack to the system.

[0053] The sensor may be a motion sensor or a proximity sensor. A software sensor is also possible. A software sensor uses a reduced algorithm to sense if there are changes in the image. A software sensor does not sense the presence of codes. Initiation of loading can thus be recognized even without a physically present sensor. The sensor may then switch to the "code recognition" mode.

[0054] Manual insertion of a classical rack comprising at least one of a sample container and a reagent container still needs to work reliably in terms of identifying codes or other features of the sample container, the reagent container or the rack at loading speeds of up to 500 mm/s.

[0055] Furthermore, insertion speeds can widely vary between different users and different loading tasks, but also during one insertion motion executed by one user. Even a retracting move can occur and needs to be detected reliably to not mix-up the correct order of identified sample containers or reagent containers on a rack during insertion.

[0056] Camera frame sizes and frame rates need to allow for taking enough image information at the maximum speed and still have enough overlap between the frames to allow software to gather all relevant information for analysis.

[0057] In addition, the exposure time for every frame needs to be short enough to 'freeze' insertion motion at maximum speed to guarantee sufficient picture quality without motion blur. For example, with a camera chip resolution that allows to reliably detect highest resolution bar codes (5 mil = 0.127 mm feature size) the camera's exposure time to avoid disadvantageous blur calculates to about 0.1 milliseconds.

[0058] Illumination has to be homogeneous and sufficiently intense. The system of the present invention satisfies the above needs.

[0059] Typically, a user needs to open a door to get access to the loading area of a system The doors are often an integrated part of the system's outer covers. Opening the door usually triggers the loading dialogue software of the graphical user interface (GUI) and starts the scanning device.

[0060] However, smaller instruments do not all have cover doors on all loading modules, i.e. systems. And opening a door does not necessarily mean that the user immediately starts with the loading procedure. The present invention may thus comprise an additional sensor, i.e. a motion sensor or a proximity sensor, that can sense approximation of a user when the loading actually takes place. This additional sensor prevents from sending a high number of camera frames to the image processing software on the data processing unit before loading has actually taken place. One embodiment of such a sensor comprised in a system is shown in Figure 2.

[0061] The sensor 210 may be a P1R (passive infrared) type that detects infrared light radiating from 'objects' in its field of view (fov) 220. The 'object' may be the user approaching the loading module 110, i.e. the system, and so triggering changes in the IR spectrum within the fov 220 of the sensor 210. The sensor 210 may be integrated in the printed circuit board assembly (PCBA) controlling the LED bar 120. As the sensing range of a PIR is typically in the meter range, the sensor 210 is oriented to 'look' downwards and the sensor fey 220 may be shielded to only allow for sensing in the loading area of the loading module 110. The optical path 130 of the camera is also shown.

[0062] In contrast to a laser bar code scanner, the information in the present invention is not analyzed within the scanning device (camera). The image frames may be forwarded to a data processing unit that may be part of an analyzer instrument, i. e. the system. An image processing software running on the data processing unit analyses the submitted frames and provides the desired data derived from detected codes and features in the desired data format.

[0063] It is possible that a defined illumination moves in accordance with the camera to adjust to the targets. The more lanes a system has the more expensive it gets to install illumination means for all lanes. From a certain number of lanes it is preferable to install illumination for only one lane that moves in accordance with the camera [0064] Illumination can be achieved with different (e. g. red) wavelengths in and out of the visual spectrum.

[0065] The camera may be equipped with different or without any filters. The camera may comprise a CCD or a CMOS chip. A colour filter may be adapted to the illumination wavelength and thus reduce or eliminate the dependency on ambient light. A polarising filter in combination with polarised light may reduce or eliminate reflexions.

[0066] Polarized light or polarizing filters are possible for illumination and detection purposes [0067] A method for identifying a sample container or a reagent container or a rack is also provided. The method comprises the following steps: focusing the camera of the system on the sample container or the reagent container or the rack being inserted into or retracted from the loading module of the system by moving the camera or the at least one minor of the system on the sledge of the system and directing the optical path of the camera to the at least one of the sample container, the reagent container and the rack, illuminating the sample container or the reagent container or the rack by the LED bar of the system, recording at least one picture by the camera, wherein the picture comprises a frame covering at least part of the sample container or the reagent container or the rack.

[0068] The method allows identifying at least one of a code, a size and/or type of at least one of a sample container, a reagent container and a rack, a fill level of a container, a size and/or type of a cap of a reagent container, a position of at least one of a sample container and a reagent container and a presence of at least one of a sample container and a reagent container.

[0069] The method may further comprise as a first step sensing an approximation of a user by the sensor of the system and thereby starting the sequence of method steps. Thus, focusing the camera of the system and all following steps may be performed after sensing an approximation of a user.

[0070] In case a code is identified, the code may be located on at least one of the sample container, the reagent container and the rack for identifying the sample container or the reagent container or the rack. The code may be located anywhere on the sample container, the reagent container or the rack as long as the setup allows the camera to focus the code.

[0071] Advantages of the present invention comprise: The design allows for the implementation of a range of cameras, lenses and illumination means from different sources and will still be able to perform within desired specifications. That simplifies the instrument integration and reduces the single source dependency risk that one classically has with highly specific laser bar code scanners.

The design supports fast manual insertion of classical sample containers and reagent containers on racks at varying (manual loading) speeds up to about 400 or 500 mm/s. Faster loading speeds are also possible. Even a retracting move (change from forward to backward motion) by the user is detectable.

The design of hardware and software allows for generating immediate feedback on reading efficiency and reading success towards user (instrument GUI) and instrument software.

The quality and intensity of the illumination for code and/or feature detection is identical on every rack insertion lane.

- The field of view and the resolution for code and/or feature detection are constant on every rack insertion lane. The way of folding the optical path between the camera and the target in combination with the focal length of the camera lens allows for adaptation of the design to smaller/wider, respectively deeper/shorter rack loading modules.

- A minimum distance between camera and target can be achieved by moving the camera or the mirror. The necessary space for the moving can easily be relocated to areas of the system that are not critical for the overall system design.

The illumination concept allows for having identical lighting properties on all rack insertion lanes. The exemplary far red LEDs (660 nm wavelength) allow for a high contrast, an illumination color that the user is accustomed to from laser bar code scanners, and, in combination with a matching filter on the camera lens, for illumination that is unsusceptible to ambient/interfering light.

The risk of total reflection artefacts and thus of data loss on the target images is practically eliminated as there is an exemplary angle (may be about 80°) between the illumination and the optical path Upgrading the image processing software will allow to maximize the bandwith of features that can be identified

EXAMPLES

Example of a system of the present invention with a moving camera [0072] In Figure 3, the camera 170 moves alongside the system to focus on the lane in which a rack 140 comprising sample containers and/or reagent containers is inserted. The lane number for the insertion is user/system predefined via a GUI loading dialogue. This allows the camera drive or the sledge motor to move the camera 170 to the desired position prior to the user's rack insertion. The double arrow in Figure 3 illustrates the moving direction. This GUI-based predefinition of the insertion lane is similar or equal to currently implemented technology.

[0073] In Figure 3, the optical path 130 of the camera 170 is only folded once via a tilted mirror 160. That implies that the travel of the camera 170 needs to be at least as long as the distance between nearest and furthest rack insertion lane where the targets are inserted.

[0074] Alternatives of this setup are possible but any alternative setup potentially increases demands for illumination or system envelope. For example, the optical path of the camera may be folded several times and the required camera travel length may be approximately only half the system width (nearest to furthest target distance). Thereby, a system with a different design (wider or shorter) can be used. However, the needs for camera positioning accuracy and precision increase with these alternative approaches.

Examples of a system of the present invention with a moving reflective mirror [0075] In Figure 4, the camera 170 stands still and the reflective mirror 410 moves and adjusts the distance to the target. Hereby the optical path 130 could of course be shaped or (multi-) folded differently, the camera 170 could e. g. be mounted vertically or in any other direction.

[0076] In Figure 5, the camera 170 is mounted behind a semi-transparent mirror 510 (beam-splitter) while moving a fully reflective mirror 410 to fold the optical path 130 and adjust the distance to the target This and the previous example would lead to shorter travel ways of the mirrors compared to the travel of the camera 170. This setup may be beneficial in case of broad and short loading bays.

[0077] Thus, the at least one mirror of the system may be a fully reflective mirror or a semi-transparent mirror. Both are alternative approaches, which can be alternatively implemented, in the system of the present invention Targets [0078] As stated above, the sample container, the reagent container and the rack are also referred to as the "target". Targets can have a wide variety in material, length, diameter, cap sizes, colors etc. A camera-based identification system of the present invention can determine more than just a labeled code on such a target [0079] Beyond simple sample containers or reagent containers, the system of the present invention can also be adapted to detect codes and features on more complex sample containers or reagent containers or all other kinds of consumables that are typically loaded into an analyzer system.

Field of view on target

[0080] In an exemplary scenario for reading a bar code on a patient sample container the CLSI standard AUT02-A2 sets the goal for the reading height. The field of view (fov) height may be 100 mm at the target position to fulfill AUTO2 and to leave room for additional features to be recognized The field of view on target 710 is illustrated in Figure 6. The target is shown as 720.

[0081] The fov on target 710 width in this example is cropped to 25 mm which is a 4:1 ratio out of the camera full chip size (that is 4:3 or 16:9 depending on camera type). With an exemplary maximum insertion speed of 500 mm/s and a frame rate of 40 frames per second the user inserts the rack with a maximum speed of 12.5 mm per frame. At a 25 mm width of the fov on target 710 the camera generates an image overlap of 100%. That is sufficient for reliable code and feature recognition, position and order detection, as well as to identify a retracting move and to generate a complete picture as an overview of the inserted rack, if desired for e. g. documentation purposes. Lens

[0082] A fast lens with a low f-number is preferable for the demands of illumination. On the opposite side, higher lens speeds drive higher cost and reduce the depth of focus. As the features that have to be detected within a frame will not all be located at the exact same focus distance, a f-2 lens may be chosen to support all opposing needs. A focal length of mm supports loading module widths of up to 300 mm, and a 35 mm focal length supports loading module widths up to 500 mm. Within limitations, the design is spatially scalable by choosing the matching focal length of the camera lens.

Illumination [0083] The exposure time for freezing the moving rack during insertion is short. The demands for illuminating the scene are correspondingly high. Figure 7 illustrates two activated LEDs of the LED bar 120 close to the rack 140 insertion position. The illumination path 810 of the two activated LEDs is indicated. The angle between the illumination center lines and the camera's optical path 130 is about 80°. This exemplary angle effectively prevents from getting total reflections and dazzle on the pictures (and so data loss).

[0084] In general, a defined number of high power LEDs are activated to illuminate the scene (for example, 2 to 3) depending on the height range of features to be identified. LEDs closer to the area of interest may be driven with less power to homogenize the illumination. Activated LEDs may not be permanently powered but pulsed at a multiple of the camera frame rate to shift the pulsing frequency out of the range of human sensory perception. E.g, at a frame rate of 40 Hz (camera takes 40 frames per second) the illumination is triggered at 2 or 3 times 40 Hz (80 Hz, 120 Hz, synchronously with the camera taking frames). Illumination and exposure are performed synchronously and the duration of illumination is longer or equal to the duration of exposure.

[0085] Thus, for example, the LEDs may only be active during the exposure time of about 0.1 milliseconds of the camera. As a result, their duty cycle is below 2% even if triggered at 120 Hz. That allows for overpowering some of the activated LEDs (the ones further away from target). With this low duty cycle, the illumination appears not too bright to the user during a manual insertion of a rack.

[0086] In combination with a b/w camera, the LEDs may be chosen to operate in the far red wavelength area at about 660 nm. In combination with a corresponding filter on the camera lens, this wavelength allows to eliminate nearly all influence of ambient lighting. Additionally it boosts contrast for b/w code and feature recognition. However, other wavelengths are also possible.

[0087] If colour detection is mandatory for the identification, the camera may be replaced by an RGB version. Polarized light and a polarizing filter on the camera lens may prevent reflection and influence of ambient lighting.

List of reference numerals loading module LED bar optical path 5 rack insertion lane protective glass tilted mirror camera sledge drive camera sledge 210 sensor

220 field of view (fov)

410 reflective mirror 510 semi-transparent mirror

710 field of view on target

720 target 810 illumination path