METHOD AND DEVICE FOR IDENTIFYING LABWARE

Abstract

In one aspect the invention relates to a computer implemented method for identifying a labware item (310, 320, 330, 340, 350), the labware item (310, 320, 330, 340, 350) comprising a first optical feature and a second optical feature, wherein the method comprises the steps of acquiring of a first image of the labware item with at least a first optical recording device (150), the first image displaying at least a portion of the first optical feature; acquiring a second image of the labware item (310, 320, 330, 340, 350) with at least a second optical recording device (160), the second image displaying at least a portion of the second optical feature; and identifying the first optical feature in the first image by using at least a first identification algorithm thereby obtaining first identification data, the first identification data encoding first information on the first optical feature and information indicative of whether at least a further identification is needed, wherein if, according to the information encoded in the first identification data, the at least further identification is not needed, the method further comprises the step of identifying the labware item by using at least the first information on the first optical feature, and wherein if, according to the information encoded in the first identification data, the at least further identification is needed, the method further comprises the steps of identifying the second optical feature in the second image by using at least a second identification algorithm thereby obtaining second identification data, the second identification data encoding information on the second optical feature; and Identifying the labware item (310, 320, 330, 340, 350) by using at least the information on the second optical feature.

Claims

1. Computer implemented method for identifying a labware item (310, 320, 330, 340, 350), the labware item (310, 320, 330, 340, 350) comprising a first optical feature and a second optical feature, wherein the method comprises the steps of: acquiring (210) of a first image (300a) of the labware item (310, 320, 330, 340, 350) with at least a first optical recording device, the first image (300a) displaying at least a portion of the first optical feature; acquiring (220) a second image (300b) of the labware item (310, 320, 330, 340, 350) with at least a second optical recording device, the second image (300b) displaying at least a portion of the second optical feature; and identifying (230) the first optical feature in the first image (300a) by using at least a first identification algorithm thereby obtaining first identification data, the first identification data encoding first information on the first optical feature and information indicative of whether at least a further identification is needed, wherein if, according to the information encoded in the first identification data, the at least further identification is not needed, the method further comprises the step of: identifying (250) the labware item (310, 320, 330, 340, 350) by using at least the first information on the first optical feature, and wherein if, according to the information encoded in the first identification data, the at least further identification is needed, the method further comprises the steps of: identifying (260) the second optical feature in the second image (300b) by using at least a second identification algorithm thereby obtaining second identification data, the second identification data encoding information on the second optical feature; and identifying (270) the labware item (310, 320, 330, 340, 350) by using at least the information on the second optical feature.

2. Method according to claim 1, wherein the method comprises the step of: determining a first region of interest in the first image (300a) by using first position information about the position of the labware with respect to a work deck, and the step of identifying the first optical feature in the first image (300a) by using the first identification algorithm is carried out by using the first region of interest, and/or wherein the method comprises the step of: determining a second region of interest in the second image (300b) by using second position information on the position of the labware with respect to the work deck, and the step of identifying the second optical feature in the second image (300b) by using the second identification algorithm is carried out by using the second region of interest.

3. Method according to either claim 1 or 2, further comprising the step of: acquiring third position information on the position of the labware item (310, 320, 330, 340, 350) with respect to the work deck by using at least a position determining algorithm, wherein the first identification algorithm processes first input data, the first input data depending on the third position information, and/or the method comprises the step of: selecting the first identification algorithm among a first pool of identification algorithms, wherein the step of selecting the first identification algorithm depends at least on the third position information.

4. Method according to any one of the preceding claims, wherein the first optical recording device is a first camera, wherein the first identification algorithm depends at least on a first set of intrinsic calibration parameters associated to the first camera, and/or at least on a first set of extrinsic calibration parameters associated to the first camera.

5. Method according to any one of the preceding claims, wherein the second optical recording device is a second camera, wherein the second identification algorithm depends at least on a second set of intrinsic calibration parameters associated to the second camera, and/or at least on a second set of extrinsic calibration parameters associated to the second camera.

6. Method according to any one of the preceding claims, wherein if, according to the information encoded in the first identification data, at least a further identification is needed, the method further comprises the step of: selecting the second identification algorithm among a second pool of identification algorithms, wherein the step of selecting the second identification algorithm depends at least on the information encoded in the first identification data.

7. Method according to any one of the preceding claims wherein the second algorithm processes second input data, the second input data encoding the first information on the first optical feature.

8. Method according to any one of the preceding claims, wherein the second identification data encodes information indicative of whether at least a further identification is needed, wherein if, according to the information encoded in the second identification data, at least a further identification is needed, the method further comprises the step of: identifying the first optical feature in the first image (300a) by using at least a third identification algorithm thereby obtaining third identification data, the third identification data encoding second information on the first optical feature, and wherein if, according to the information encoded in the second identification data, at least a further identification is needed, the step of identifying the labware item (310, 320, 330, 340, 350) by using at least the information on the second optical feature is performed by using the second information on the first optical feature.

9. Method according to claim 8, wherein the third algorithm processes third input data, the third input data encoding the first information on the first optical feature and/or the information on the second optical feature.

10. Method according to any one of the preceding claims, wherein the first optical feature and/or the second optical feature comprise an indicium, an ideogram, a pictogram, a set of alphanumeric characters, a texture pattern, a hole pattern and/or a color; and/or wherein the labware item (310, 320, 330, 340, 350) comprises a plate, a tip, a tube, a reservoir, a tip box, a height adapter, a reservoir rack and/or a tube rack.

11. Method according to any one of the preceding claims, wherein if, according to the information encoded in the first identification data, at least a further identification is needed, the step of identifying the labware item (310, 320, 330, 340, 350) by using at least the information on the second optical feature is performed by using the first information on the first optical feature.

12. A data processing system (100) comprising a first optical recording device, a second optical recording device, and processing means configured to perform the method according to any one of the claims 1 to 11.

13. An automated laboratory system comprising a first optical recording device (150), a second optical recording device (160), processing means configured to perform the method according to any one of the claims 1 to 11, and a work deck for positioning a labware item (310, 320, 330, 340, 350).

14. A computer program product comprising instructions which, when the program is executed by a system according to claim 12 or 13, cause said system to carry out the method according to any one of the claims 1 to 11.

15. A computer-readable storage medium comprising instructions which, when executed by a system according to claim 12 or 13, cause said system to carry out the method according to any one of the claims 1 to 11.

Description

[0135] Exemplary embodiments of the invention are described in the following with respect to the attached figures. The figures and corresponding detailed description serve merely to provide a better understanding of the invention and do not constitute a limitation whatsoever of the scope of the invention as defined in the claims. In particular:

[0136] FIG. 1 is a schematic representation of an embodiment of the data processing system according to the present invention;

[0137] FIG. 2 is a flow diagram of the operation of a first embodiment of the method of the present invention;

[0138] FIG. 3a, 3c are schematic representations of the first image acquired by using the embodiment of the data processing system according to the present invention;

[0139] FIG. 3b, 3d are schematic representations of the second image acquired by using the embodiment of the data processing system according to the present invention; and

[0140] FIG. 4 is a flow diagram of the operation of a second embodiment of the method of the present invention.

DETAILED DESCRIPTION OF EMBODIMENTS

[0141] FIG. 1 is a schematic view of a first embodiment of the data processing system (DPS) 100 according to the present invention. Said data processing system 100 may comprise a computing device or a cluster thereof. The data processing system 100 comprises a processing element 110 and storage means 120, which are in data communication with one another. The processing element 110 may consist of or comprise a CPU and/or a GPU and comprises several modules 111-116 configured to perform the steps of the method of the present invention.

[0142] The storage means 120 may comprise volatile primary memory 121 (e.g. a RAM, a DRAM, a SRAM, a CPU cache memory or the like) and/or non-volatile primary memory 122 (e.g. a ROM, a PROM, an EPROM or the like). In particular, the volatile primary memory may consist of a RAM. For instance, the volatile primary memory 121 temporarily holds program files for execution by the processing element and related data and the non-volatile primary memory 122 may contain bootstrap code for the operating system of the DPS 100.

[0143] The storage means 120 may further comprise a secondary memory 123, which may store the operating system and/or the instructions of the algorithms used to carry out the method of the present invention. Moreover, the secondary memory 123 may store a computer program product comprising instructions which, when the computer program product is executed by the DPS 100, cause the DPS 100 to carry out the method according to the present invention.

[0144] The secondary memory 123, the primary memories 121, 122, and the processing element 110 need not be physically housed within the same housing and may instead be spatially separated from one another. In particular, the secondary memory 123, the primary memories 121, 122, and the processing element 110 may be spatially separated from one another and may exchange data with one another via wired and/or wireless media (not shown).

[0145] The DPS 100 may comprise an input/output (I/O) interface 140 which allows the DPS 100 to communicate with input/output devices (e.g. displays, keyboards, touchscreens, printers, mice, cameras, or the like). The DPS 100 may further comprise a network interface controller (NIC) 130 configured to connect the DPS 100 with a suitable network (not shown). According to the present invention, a suitable network may for instance be an intranet, the internet or a cellular network.

[0146] The data processing system 100 comprises a first optical recording device and the second optical recording device in the form of a first digital camera 150 and a second digital camera 160, respectively. In particular, the first camera 150 and the second camera 160 are a greyscale and a polychrome camera, respectively. The first camera 150 and the second camera 160 are configured to acquire the first image and the second image, respectively, and may be a photo camera and/or a video camera. As shown in FIG. 1, the first camera 150 and/or the second camera 160 may be connected to the processing element 110 via the I/O interface 140. For example, the first camera 150 and/or the second camera 160 may be connected wirelessly to the I/O interface via the NIC 130. The first camera 150 and/or the second camera 160 may be intelligent devices with their own memory for storing associated instructions and data for use with the I/O interface 140 or peripheral devices.

[0147] The processing element 110 comprises several modules 111 to 116 configured to carry out the method of the present invention. In particular, the first acquisition module 111 is configured to operate the first camera 150 to acquire the first image. The second acquisition module 112 is configured to operate the second camera 160 to acquire the second image. The processing element 110 comprises a first identification module 113 configured to identify the first optical feature in the first image by using the first identification algorithm. The second identification module 114 is configured to identify the labware item by using the first information on the first optical feature. The processing element 110 further comprise a third identification module 115 configured to identify the second optical feature in the second image by using at least a second identification algorithm. The fourth identification module 116 is configured to identify the labware item by using the information on the second optical feature. The first identification module 113 and the third identification module 115 may be the same module. In particular, the second identification module 114 and the fourth identification module 116 may be the same module. The processing element 110 may further comprise an evaluation module (not shown) configured to establish whether, according to the information encoded in the first identification data, a further identification is needed.

[0148] For example, the DPS 100 may be a computer system including the first camera 150 and the second camera 160 and, more particularly, may be a smartphone, a desktop computer, a tablet, a laptop or the like. Moreover, the DPS 100 may be the ALS according to the present invention, in particular an automatic pipetting system. In this case, the DPS 100 comprises a work deck (not shown) for positioning one or more labware items and/or a pipetting head (not shown) for liquid transfer. The pipetting head may be movable with respect to the work deck by means of servo and/or stepper motors.

[0149] FIG. 2 is a flow diagram 200 of the operation of a first embodiment of the method according to the present invention. In particular, the first embodiment of the method according to the present invention may be performed by the DPS 100 described above and schematically depicted in FIG. 1.

[0150] The first embodiment of the method carries out the identification of a 1000 μl tipbox with filter. This identification is carried out by using the first and the second optical feature. The first optical feature is an L-shaped hole pattern 341 comprising four holes and the second optical feature is the color of the 1000 μl tipbox with filter, e.g. the color cyan. For the sake of discussion, it is assumed that the 1000 μl tipbox with filter is the only labware item comprising these two optical features.

[0151] At step 210, the DPS 100 acquires the first image 300a, schematically represented in FIG. 3a and FIG. 3c. The first image 300a is acquired by using the first camera 150 and, hence, is a greyscale image. As shown in FIG. 3a, the first image 300a displays a top view of the work deck 370 of the DPS 100. The work deck 370 comprises a waste container 350 and four regions 371-374 for positioning labware items.

[0152] A first labware item 310, which is a 96 MTP Plate, is positioned on the first region 371. The first labware item 310 comprises ninety-six wells 312 distributed in twelve columns and eight rows. The first labware item 310 comprises two sets of alphanumeric characters 311, 313. The first set of alphanumeric characters 321 forms a column depicting the letters A, B, C, D, E, F, G and H. The second set of alphanumeric characters 323 forms a row depicting the numbers from one to twelve. The wells 312 of the first labware item 310 contain a first compound.

[0153] A second 320 labware item and a third labware item are positioned on the second region 372. The second labware item 320 is a 96 MTP Plate comprising 96 wells 322 distributed in 12 columns and 8 rows. The second labware item 320 comprises two sets of alphanumeric characters 321, 323. The first set of alphanumeric characters 321 forms a column depicting the letters A, B, C, D, E, F, G and H. The second set of alphanumeric characters 323 forms a row depicting the numbers from one to twelve. The wells 322 of the second labware item 320 contain a second compound. The third labware item is a 55 mm height adapter, onto which the second labware 320 item is arranged. In particular, FIG. 3a depicts the five pins 331-335 of the height adapter.

[0154] A fourth labware item 340, which is a 1000 μl tipbox with filter, is positioned on the third region 373. The fourth labware item 340 comprises ninety-six tips 342 distributed in twelve columns and eight rows and an L-shaped hole pattern 341 comprising four holes. A fifth labware item 360, is positioned on the fourth region 374. The fifth labware item 360 is a reservoir rack comprising a 100 ml tube 364, a 30 ml tube 363, a first reservoir rack module 362, and a second reservoir rack module TC 361. The first reservoir rack module 362 comprises four reaction vessels having a diameter of 16 mm and/or the second reservoir rack module 361 comprises two reaction vessels having a diameter of 29 mm. in particular, the first reservoir rack module 362 and/or the second reservoir rack module 361 may be a temperature controlled by means of a thermal module (not shown).

[0155] At step 220, the DPS 100 acquires the second image 300b, schematically represented in FIG. 3b and FIG. 3d. The second image 300b is acquired by using the second camera 160 and, hence, is a polychrome image. As shown in FIG. 3b, the second image 300a displays a top view of the work deck 370 of the DPS 100, said top view being substantially identical to the one displayed in the first image 300a, cf. FIG. 3a. The second image 300b is acquired by using a polychrome camera and, hence, the second image 300b differs from the first one 300a at least in that the former image 300a displays the colors of the labware items. In particular, in FIG. 3b, the color of the first 310, the second 320, the third labware item and the fourth 340 labware item are schematically represented by hatched regions. For example, the first labware item 310 is green, the second labware item 320 is yellow, and the fourth labware item 340 is cyan. Both the first and the second labware item are a 96 MTP plate. However, they contain different compounds and, hence, are distinguished from one another by their color.

[0156] The first image 300a and/or the second image 300b may be stored in the primary and/or secondary memory of the DPS 100 and may be accessed by the processing element 110 to identify the first optical feature and/or the second optical feature, respectively.

[0157] At step 230, the DPS 100 identifies the first optical feature, i.e. the L-shaped hole pattern 341, in the first image 300a by using the first identification algorithm. In particular, the first identification algorithm processes first input data that comprise the location and the intensity of the pixels of the first image 300a. The first identification algorithm identifies the first optical feature by checking whether the L-shaped hole pattern 341 comprising four holes is displayed in the first image 300a and by providing an estimate of the location of said hole pattern 341 in the first image 300a. In particular, the first identification algorithm may comprise machine learning algorithm, e.g. an ANN.

[0158] The first identification algorithm generates the first identification data as output, said data comprising information on the first optical feature. The information on the first optical feature specifies that, with a first specified probability, an L-shaped hole pattern 341 is displayed at a specified location in the first image 300a and with a specified orientation with respect to a given direction. In FIG. 3c, the specified location is marked by the dotted box 380 and the given direction is denoted by the dot-dashed line 390. In particular, as shown in FIG. 3c, the specified orientation of the L-shaped hole pattern 341 is such that the longer arm of the L formed by the hole pattern 341, i.e. the arm comprising three holes, is substantially parallel to the given direction 390.

[0159] At step 240, it is established whether, according to the information encoded in the first identification data, a further identification is needed. This may be achieved by comparing the first specified probability with a given threshold. The given threshold may, for instance, be comprised between 0.8 or 0.9. Hence, in this embodiment, the first specified probability encodes the information indicative of whether at least a further identification is needed.

[0160] If the probability is larger than the given threshold, the DPS 100 carries out the first labware identification algorithm which processes input data and generates the first labware identification data (step 250). For example, in this case, the input data of the first labware identification algorithm comprises information specifying that, with the first specified probability, the L-shaped hole pattern 341 is displayed at the specified location in the first image 300a and with a specified orientation with respect to the given direction 390.

[0161] The first labware identification algorithm comprises instructions that, when executed by the processing element 110, cause the processing element to access a lookup table to assess what is the labware item associated with the first optical feature, i.e. the labware item comprising an L-shaped hole pattern 341 with four holes. The lookup table associates the first optical feature with the 1000 μl tipbox with filter, thus the first labware identification data comprises information specifying that the identified labware item is a 1000 μl tipbox with filter 340.

[0162] The first labware identification algorithm comprises instructions that, when executed by the processing element 110, cause the processing element to estimate the location of the 1000 μl tipbox with filter 340 on the work deck 370. Said estimate may be obtained by using information indicative of the shape and the features of the identified labware item, and of the specified location and orientation of the L-shaped hole pattern 341 in the first image 300a. Hence, the first labware identification data comprises information specifying that the identified labware item is located in the fourth region 374 of the work deck 370.

[0163] In particular, the information indicative of the shape and the features of the identified labware item comprises information specifying that the identified labware item as displayed in the first image 300a has a rectangular shape with specified dimensions and ninety-six wells. Moreover, the 1000 μl tipbox with filter 340 comprises a L-shaped hole pattern 341 with specified dimensions. The hole pattern 341 of the 1000 μl tipbox with filter 340 is oriented in such a way that the longer arm of the “L” formed by the hole pattern is substantially parallel to the shorter side 343 of the 1000 μl tipbox with filter 340, cf. FIG. 3a or 3c.

[0164] The first labware identification algorithm may use the extrinsic calibration parameters, the intrinsic calibration parameters of the first camera, and the dimensions of a 1000 μl tipbox with filter 340 to estimate the size of the region of the first image 300a, that displays the 1000 μl tipbox with filter 340.

[0165] If, instead, the probability is lower than the given threshold, the DPS 100 identifies the second optical feature in the second image 300b by using the second identification algorithm (step 260). In particular, this embodiment may comprise the step of selecting the second identification algorithm among the second pool of identification algorithms. The selection of the second identification algorithm depends at least on the information encoded in the first identification data, i.e. on the information specifying that the first optical feature is, with the first specified probability, an L-shaped hole pattern 341. As the only labware item comprising an L-shaped hole pattern 341 is a 1000 μl tipbox with filter 340, the second identification algorithm selected by the DPS 100 assesses whether the second image 300b comprises regions with pixels having intensity corresponding to the color cyan.

[0166] More specifically, the second identification algorithm processes second input data that comprise the location and the intensity of the pixels of the second image 300b. The second identification algorithm comprises instructions which, when executed by the processing element 110, causes the DPS 100 to assesses whether the second image 300b comprises regions with pixels having intensity corresponding to the color cyan and the location of said regions. In particular, said assessment is carried out by locating the pixels with intensity falling within a range of intensities that corresponds with the color cyan. For example, if the intensity is expressed in terms of the RGB color model, the range of intensities may comprise the intensities having first RGB value comprised between 0 and 100, second RGB value comprised between 200 and 255 and second RGB value comprised between 200 and 255. Alternatively, if the intensity is expressed in terms of the RGB color model, the intensity range associated to the color cyan may be expressed by using the HSL or HSV representation.

[0167] The second identification algorithm generates the second identification data as output, said data comprising information on the second optical feature. The information on the second optical feature specifies that, with a second specified probability, the intensity of the pixels of a rectangular region of the second image 300b corresponds to the color cyan, said rectangular region being located at a specified location of the second image 300b. In FIG. 3d, the rectangular region of the second image 300b is marked by a dotted box 385.

[0168] The second input data may further comprise information indicative of the specified location of the L-shaped hole pattern 341 in the first image 300a and the second identification algorithm may detect the rectangular region 385 by using said information. For instance, the second identification algorithm may use the intrinsic and extrinsic parameters of the first 150 and the second 160 camera to estimate, given the specified location of the hole pattern 341 in the first image 300a, the corresponding location of said pattern 341 in the second image 300b. The rectangular region 385 may thus be detected by analyzing the intensity of the pixels located in a region comprising the corresponding location of the L-shaped hole pattern 341 in the second image 300b.

[0169] At step 270, the labware item is identified by using the first information on the first optical feature 341 and the information on the second optical feature. More specifically, the DPS 100 carries out the second labware identification algorithm which processes input data and generates the second labware identification data. For example, the second labware identification algorithm is a decision tree and the input features are the first and the second specified probabilities, the size, specified location, and specified orientation of the optical feature 341 in the first image 300a, and the size, specified location, and orientation of the rectangular region in the second image 300b. In this case, the second labware identification algorithm assesses, given the input features, whether the labware item is a 1000 μl tipbox with filter 340 and, if this is the case, its location on the work deck 370. Thus, in this embodiment, the second labware identification data may specify that the labware item is a 1000 μl tipbox with filter 340 located in the fourth region 374 of the work deck 370.

[0170] The first identification data, the second identification data, the first labware identification data and/or the second labware identification data may be stored in the primary memory and/or secondary memory of the DPS 100.

[0171] A further embodiment of the method according to the present invention may comprise the steps 210 to 270 of the first embodiment described above. This former embodiment differs from the latter one in the order, according to which the steps 210 to 270 are carried out. in particular, in the further embodiment, the step 220 of acquiring the second image 330b may be carried either before the step 210 of acquiring the first image 330a or after the step 230 of identifying the first optical feature 341 by using the first identification algorithm.

[0172] FIG. 4 is a flow diagram 400 of the operation of a second embodiment of the method according to the present invention which, for instance, may be performed by the DPS 100 described above and schematically depicted in FIG. 1. The second embodiment of the method carries out the identification of the labware item located on the second region of the work deck of the DPS 100.

[0173] At step 405, the DPS 100 acquires the first image 300a by using the first camera 150. The first image 300a is a greyscale image and is substantially identical to the homonymous image acquired by carrying out the first embodiment of the method of the present invention. Hence, the first image 300a acquired by carrying out the step 405 is schematically represented in FIG. 3a and FIG. 3c. In particular, the first image 300a displays a top view of the work deck 370 of the DPS 100. The work deck 370 comprises a waste container 350 and four regions 371-374 for positioning labware items.

[0174] At step 410, the DPS 100 acquires the second image 300b by using the second camera 160. The second image 300b is a polychrome image and is substantially identical to the homonymous image acquired by carrying out the first embodiment of the method of the present invention. Hence, the first image 300a acquired by carrying out the step 410 is schematically represented in FIG. 3b and FIG. 3d.

[0175] At step 415, the DPS 100 determines the first ROI in the first image 300a by using first position information. Said information specifies that the labware item to be identified is the one positioned on the second region 372 of the work deck 370. The first ROI is a rectangular region of the first image 300a displaying the second region 372 of the work deck 370.

[0176] The first ROI is obtained by using the first ROI determining algorithm by processing the pixels of the first image 300a and the first position data encoding the first position information. For example, the first ROI determining algorithm uses the intrinsic and extrinsic calibration parameters associated with the first camera 150 to detects the rectangular region of the first image 300a that displays the second region 372 of the work deck 370.

[0177] At step 420, the DPS 100 acquires the third position information on the position of the to-be-identified labware item with respect to the work deck 370. Said information is generated by using at the position determining algorithm. Said algorithm processes position determining input data, which comprise the location and the intensity of the pixels comprised in the first ROI.

[0178] The position determining algorithm comprises an ANN that processes the position determining input data to determine whether the first ROI displays a height adapter by detecting at least some of the pins of said height adapter. As shown in FIG. 3a or FIG. 3c, the first ROI displays the pins 331-335 of a 55 mm height adapter. Hence, the ANN detects said pins 331-335 and provides an estimate of their location in the first ROI and an estimate of the height of the height adapter. In particular, said height is estimated to be equal to 55 mm.

[0179] The position determining algorithm comprises instructions which, when executed by the processing element 110, cause the DPS 100 to acquire third position information by estimating the location of the labware item to be identified. Said estimate is carried out by using the locations of the pins 331 to 335 in the first ROI. In particular, the third position information specifies that the labware item to be identified is comprised in a third ROI of the first image 330a. In FIG. 3c, the third ROI is marked by the dashed square 399. The third position information further specifies the estimated height of the height adapter displayed in the first ROI.

[0180] At step 425, the first identification algorithm is selected among a first pool of identification algorithms. This selection depends on the location of the first ROI in the first image 300a and on the third position information, in particular on the estimated height of the height adapter displayed in the first ROI. For the sake of discussion, it is assumed that only two labware items may be arranged on a height adapter of the estimated height, namely a 96 MTP Plate or a 1000 μl tipbox with filter.

[0181] Accordingly, the selected identification algorithm is a confidence-weighted classifier performing a multiclass classification to detect an L-shaped hole pattern comprising four holes and/or two sets of alphanumeric characters, the first one forming a column depicting the letters from A to H, the second one forming a row depicting the numbers from one to twelve. Moreover, this algorithm may be optimized to carry out a multiclass classification on the scene displayed in the first ROI, e.g. to classify the objects located in the second region 372 of the work deck. In particular, when identifying the second optical features, the first identification algorithm takes into account the perspective of the image, which depends, inter alia, on the position of the second region 372 with respect to the first camera. The first identification algorithm may also take into account the illumination conditions, that, typically, characterize the second region 372 of the work deck 370.

[0182] At step 430, the DPS 100 identifies the first optical feature by using the first identification algorithm. In particular, the first identification algorithm processes first input data that comprise the location and the intensity of the pixels of the third ROI 399. Said algorithm identifies the first optical feature by checking whether the third ROI 399 displays an L-shaped hole pattern comprising four holes or two sets of alphanumeric characters, the first one forming a column depicting the letters from A to H, the second one forming a row depicting the numbers from one to twelve.

[0183] The first identification algorithm generates the first identification data as output, which comprise information on the first optical feature. In this case, the information on the first optical feature may specify that, with a third specified probability, the third ROI 399 displays two sets of alphanumeric characters, the first set 321 forming a column depicting the letters from A to H and the second set 323 forming a row depicting the numbers from one to twelve. The information on the first optical feature may further specify that, with a fourth specified probability, the third ROI 399 does not display an L-shaped hole pattern comprising four holes. Furthermore, the first information on the first optical feature specifies that, with a fifth specified probability, the first optical feature consists of two sets of alphanumeric characters, the first set 321 forming a column depicting the letters from A to H and the second set 323 forming a row depicting the numbers from one to twelve.

[0184] In particular, the fifth specified probability is equal to 0.98 and indicates that the requirements on the accuracy of the identification of the first optical feature are met. In this case, the first information on the first optical feature is reliable but inconclusive. In particular, said information allows for concluding that the labware item 320 is a 96 MTP plate, i.e. that the said item belongs to the class of 96 MTP plates. The first information on the first optical feature, however, is not sufficient to identify which compound is contained in said labware item. In order to identify the compound, the second optical feature, i.e. the color, of the labware item 321 has to be identified.

[0185] Hence, in this embodiment, the first information on the first optical feature specifies that at least a further identification is needed. Hence, at step 435, it is established that, according to the first information on the first optical feature, a further identification is needed and, thus steps 445, 450 and 455 have to be carried out.

[0186] At step 445, the DPS 100 selects the second identification algorithm among the second pool of identification algorithms. The selection of the second identification algorithm depends at least on the first information on the first optical feature. As the only class of labware item comprising the first optical feature is the class of 96 MTP plates, the second identification algorithm selected by the DPS 100 identifies the average color of a 96 MTP plate displayed in the third ROI 399.

[0187] At step 450, the DPS 100 identifies the second optical feature in the image 300b by using the second identification algorithm. The second identification algorithm processes second input data that comprise the location and the intensity of the pixels of the second image 300b and information indicative of the location of the third ROI 399. For instance, the second identification algorithm may use the intrinsic and extrinsic parameters of the first 150 and second 160 camera to estimate, given the specified location of the third ROI 530 in the first image 300a, a corresponding fourth ROI in the second image 300b. In FIG. 3d, the fourth ROI is marked by a dot-dashed box 395 and displays the 96 MTP Plate 320.

[0188] The second identification algorithm comprises instructions which, when executed by the processing element 110, causes the DPS 100 to compute the average intensity of the pixels of the fourth ROI 395 and to identify the color of the labware item. In particular, the color identification is carried out by assessing whether the average intensity falls within the intensity range associated with the color yellow or within the range associated with the color green. For example, if the intensity is expressed in terms of the RGB color model, the intensity range associated with the color green comprises the intensities having first RGB value comprised between 0 and 100, second RGB value comprised between 200 and 255 and third RGB value comprised between 0 and 100. The intensity range associated with the color yellow may comprise the intensities having first RGB value comprised between 230 and 255, second RGB value comprised between 230 and 255 and third RGB value comprised between 0 and 100. If the intensity is expressed in terms of the RGB color model, the intensity range associated to the color green and/or the intensity range associated to the color yellow may also be expressed by using the HSV or the HSL representation.

[0189] Alternatively, or in conjunction with the above, the second identification algorithm may comprise an ANN and/or a SVM which process the pixels of the fourth ROI 395 to identify the color of the labware item.

[0190] The second identification algorithm generates the second identification data as output, which comprise information on the second optical feature. As the fourth ROI 395 displays a yellow 96 MTP Plate 320, the information on the second optical feature specifies that, with a sixth specified probability, the color of the labware item displayed in the fourth ROI 395 is the color yellow.

[0191] At step 4550, the labware item is identified by using the first information on the first optical feature 321, 323 and the second optical feature. More specifically, the DPS 100 carries out the second labware identification algorithm which processes input data and generates the second labware identification data. For example, in this case, the input data of the second labware identification algorithm comprises information encoded in the first information on the first optical feature 341 and the information on the second optical feature.

[0192] In particular, the second labware identification algorithm is a decision tree. In this embodiment, the second labware identification data may specify that the labware item located in the second region 372 of the work deck 370 is a yellow 96 MTP plate 320 arranged on a 55 mm height adapter 331 to 335 and that the well of the 96 MTP plate 320 contains the second compound.

[0193] If, at step 430, the first identification algorithm generated the first identification data specified that the first optical feature was an L-shaped hole pattern, the step 440 would have been carried out instead of the steps 445, 450 and 455. In this case, the labware item would have been identified to be a 1000 μl tipbox with filter, which is the only labware item comprising an L-shaped hole pattern.

[0194] Wherever not already described explicitly, individual embodiments, or their individual aspects and features, described in relation to the drawings can be combined or exchanged with one another without limiting or widening the scope of the described invention, whenever such a combination or exchange is meaningful and in the sense of this invention. Advantages which are described with respect to a particular embodiment of present invention or with respect to a particular figure are, wherever applicable, also advantages of other embodiments of the present invention.