A method of, and a system and a computer program for analysing a baked goods item

Abstract

A method of analysing a crumb structure of a baked goods item, comprising steps of: receiving at a mobile device, visual data of the baked goods item, including real-time distance measurements of the distance from the mobile device to the baked goods item, 5 wherein the visual data has been collected in a light controlled environment; segmenting the received visual data to identify a contour of the baked goods item; and segmenting the area inside the contour of the baked goods item; estimating a plurality of crumb structure properties of the baked goods item, using the results of the previous step, by separating regions within the baked goods item based on pixel intensity wherein regions of lower 0 intensity correspond to open spaces or voids (i.e., pores, holes), and regions with higher intensity correspond to the crumb.

Description

[0001] A METHOD OF, AND A SYSTEM AND A COMPUTER PROGRAM FOR ANALYSING A BAKED

[0002] GOODS ITEM

[0003] Technical Field of the Invention

[0004] The present invention relates to a method of, and a system for, analysing a baked goods item.

[0005] 5 Background to the Invention

[0006] Baked goods are foods made from dough or batter and cooked by baking, a method of cooking food that uses prolonged dry heat, normally in an oven. The item most baked is bread but many other types of foods are baked as well.

[0007] It will be appreciated by those in the industry that various standard parameters exist which0 are used to measure the quality of baked goods, e.g., the volume and structure of the product.

[0008] Within food quality control processes, samples of baked goods are often routinely taken and monitored by food quality control specialists & technologists. In this regard, they will carry out a scientific analysis on the food samples to determine whether the food quality is of sufficient standard. This helps maintain critical consistency in the production of bread and other baked items.

[0009] One of the existing tools for measuring food quality is the Calibre Control™ C-Cell™ which provides a crumb structure data analysis for bread, cake and other structurised products. The C-Cell™ system uses an entry-level digital imaging system to take a photo of a slice of the baked goods within a light-controlled environment (generally, within the machine itself) and0 then performs a calculation of the data on a computer connected to the device. The C-Cell™ system allows technical, production and engineering personnel to set their own key performance indicators, using objective data. In this manner, a qualitative determination of food quality can be made. The tool requires a specific and constant focus distance between the camera and the baked goods item for accurate operation. This causes meticulous sample preparation, requiring baked goods to be sliced to a precise thickness within a laboratory setting. For industrially sliced bread, the system allows for slice positioning within the apparatus in discrete increments of 5 mm. This introduces a standard deviation for slices that do not correspond to a multiple of 5 mm, as such slices may not be within optimal camera focus, leading to inaccuracies in size conversion. These factors often result in operational inefficiencies and increased deviation in analytical outcomes. All of the above leading to an increase in the size, complexity, and cost of the device. In addition, it will be appreciated by those in the art, that industrial equipment of this nature is not portable and requires specialized technical training to use effectively.

[0010] US2022012467 discloses a multi-sensor analysis of food products, such as baked goods, including utilising a variety of existing detection methods such as a standard camera on a mobile phone, a LiDAR (light detection and ranging) sensor and artificial intelligence engine. This document discloses an investigation into the calorific content of food through an estimation of the volume and composition of the food. More particularly, the described technology includes receiving a 3 -dimensional (3D) image, identifying a food item in the 3D image, determining a volume of the identified food item based on the 3D image, and estimating a composition of the identified food using a millimeter-wave radar.

[0011] However, the cited prior art does not provide a means of analysing the crumb structure within a food item, such as within baked goods. Such an analysis of crumb structure would provide a determination of the key crumb quality attributes of the baked goods. For example: porosity level, cell count, volume of the slice, cell wall thickness, concavity level etc. This could form an important part of a multi-faceted food quality assessment program.

[0012] As such, a need exists for a means of analysing crumb structure within this context.

[0013] Further, one of the challenges in analysing the crumb structure of industrially sliced baked goods is the uneven surface of the baked goods item, which may arise from certain processing conditions or recipe factors, such as elevated internal temperatures during slicing or variations in ingredient quality, such as poor flour quality. Existing methods of crumb structure analysis do not adequately address these surface irregularities. It is an object of the invention to provide a method of, and a system for, analysing a baked goods item, which provides the advantages and addresses some of the issues and deficiencies described above.

[0014] Summary of the Invention

[0015] According to a first aspect of the invention, there is provided a method as claimed in claim 1, and corresponding system and computer program as claimed.

[0016] The present invention eliminates the need to maintain a fixed distance between a camera device and the baked goods item during visual data capture and therefore allows for a portable device to be used for measuring crumb structure characteristics of baked goods items. For example, the handheld use of a smartphone to capture visual data of the baked goods item for subsequent analysis, can be used, as described below. This enables analysis to occur at the required location and enhances operational efficiency by accelerating the speed of analysis.

[0017] Preferably, the invention also provides that the light controlled environment includes a light source configured to emit minimal or no radiation in the near-infrared range, and the light source is also configured to emit light parallel to the surface of the baked goods item wherein the baked goods item is located in a lightbox, wherein the light source is located inside the lightbox in such a configuration that the light source emits light from each of a plurality of sides of the baked goods item and across the height of the baked goods item.

[0018] This addresses the problem mentioned above of the uneven surfaces and deals with the surface irregularities by emitting the light from each side of the analysed baked goods item, that is placed inside the lightbox, and across baked goods item whole height. This allows even light distribution despite the quality of the baked goods item cut and / or presence of irregularities on the surface of the baked goods item. Brief Description of the Drawings

[0019] These and other features of this invention will become apparent from the following description of one example described with reference to the accompanying drawings in which:

[0020] Figure 1 shows a positioning of the smartphone during analysis, including the method of distance measurement;

[0021] Figure 2 shows a design of the lightbox device, in accordance with aspects of the invention;

[0022] Figure 3 shows light distribution from the light panel that is forming the edges of the lightbox device, and images of bread slices;

[0023] Figure 4 shows an application view of the functional blocks of a system in accordance with a preferred embodiment;

[0024] Figures 5a shows the smartphone device and the view of an open mobile application screen before capturing the visual data of the baked goods item;

[0025] Figure 5b illustrates an open mobile application screen with identified contour of the baked goods item using a SOD methodology, in particular a U2-Net model;

[0026] Figure 6 shows an example of a binary representation of the image after it is processed by software;

[0027] Figure 7 shows an application view with the results of the analysed product;

[0028] Figure 8 shows a flow chart displaying a method of analysing a baked goods item, in accordance with aspects of the invention; and Figure 9 shows a computer within which a set of instructions, for causing the computer to perform any one or more of the methodologies described herein, in accordance with aspects of the invention.

[0029] Detailed Description of Preferred Embodiments

[0030] The following description of the invention is provided as an enabling teaching of preferred embodiments of the invention. Those skilled in the relevant art will recognise that many changes can be made to the embodiment described, while still attaining the beneficial results of the present invention. It will also be apparent that some of the desired benefits of the present invention can be attained by selecting some of the features of the present invention without utilising other features. Accordingly, those skilled in the art will recognise that modifications and adaptions to the present invention are possible and can even be desirable in certain circumstances and are a part of the present invention. Thus, the following description is provided as illustrative of the principles of the present invention and limitation thereof.

[0031] Figure 1 illustrates the positioning of a smartphone 102 in relation to a lightbox 104 during the capture of visual data related to a baked goods item 106, such as a slice of bread, for analysing the crumb structure of the baked goods item 106, as will be described below. In an embodiment of the invention, the visual data is provided in the form of one or more image(s), video(s) and / or 3-dimensional (3D) image(s) of the baked goods item. Illustration 150 depicts a typical configuration of sensors on modem smartphones. Part of the visual data, such as image(s) and / or video(s), is captured by the smartphone camera(s) 107. In an embodiment of the invention, the 3 -dimensional image(s) are provided in the form of 3- dimensional image data, such as computer vision data. In another embodiment, the 3- dimensional image data is provided in the form of 3-dimensional mesh data or point cloud data. In embodiments, the 3 -dimensional image data, which includes distance information as part of the visual data, is captured using a ranging light detection and ranging (LiDAR) sensor 108. LiDAR pulses 108a are emitted from a LiDAR sensor 108 of the smartphone 102 and reflect off the surface of the baked goods item, returning 108b to the LiDAR sensor 108 of the smartphone 102 providing real-time data regarding the distance from the smartphone 102 to the baked goods item 106. The smartphone may be hand held by the user during the analysis. In the present embodiment, the distance from the smartphone 102 to the baked goods item 106 is determined by analysing the LiDAR data received from a rectangular area located at the center 110 of the baked goods item 106. The real-time distance data, in conjunction with known camera properties, is utilised to calculate the spatial resolution. Spatial resolution is best defined as the physical dimension that represents a pixel of the image. This is used for estimating crumb structure attributes, such as cell wall size, cell diameter, and other relevant characteristics.

[0032] A process of spatial resolution calculation is carried out in two steps: (a) determining the sensors true height size (hereinafter referred to as "THS"), and (b) based on the image resolution, calculating the spatial resolution once the THS value is known. A sensor's true height size refers to the actual vertical field of view distance captured by the smartphone camera, as determined by the sensor's height.

[0033] Distance to object, mm x Sensor heiqht, mm

[0034] THS,mm = - - - - - — - - -

[0035] Camera focal length, mm

[0036] T HS, mm

[0037] Spatial resolution, mm / pixel = -; ;- — - —

[0038] Vertical resolution of the image, pixels

[0039] For example, the LiDAR sensor has received data indicating that the distance between the smartphone 102 and the baked goods item 106 is 250 mm. The smartphone camera lens has a focal length of 6.7 mm, and the height of the sensor is 7.3 mm. The image resolution is 2800 by 2800 pixels. The following calculation is then performed:

[0040] 250 x 7.3

[0041] THS, mm = = 272

[0042] 6.7

[0043] 272

[0044] Spatial resolution, mm / pixel — — = 0.097 2800

[0045] The spatial resolution will be utilised in (as will be described below) the third software module 414, as shown in Figure 4, for calculating specific crumb structure characteristics such as cell wall size, cell diameter, and others. The results of this calculation will be presented later, as depicted in Figure 7.

[0046] In the present embodiment, the LiDAR sensor emits waves at a wavelength of approximately 940 nm. The lightbox 104 is equipped with an electroluminescent (EL) panel that emits cool white light with wavelength ranging from 400 to 650 nm, which produces minimal nearinfrared radiation. This configuration ensures minimal interference with the LiDAR sensor's operation, thereby optimising the sensor's performance in capturing accurate real-time distance data. In some embodiments filters for LiDAR sensor could be utilised to accommodate any light source within the lightbox. In yet another embodiment a monochrome light could be used as the light source of the lightbox.

[0047] Figure 2 shows a lightbox device 200, according to the embodiment of the invention. The lightbox assembly features a light emitting EL panel 202 that ensures even (uniform) illumination inside the lightbox. The EL panel 202 is emitting the light parallel to the surface of the baked goods item placed for the analysis creating accurate shadows in the voids / empty spaces across the whole surface. The EL panel is made of a flexible material and preferably consists of a single piece. EL panel is re-attachable to the base of the lightbox 204 and preferably assembles to the base by a Velcro™ connection 206a and 206b. The additional Velcro™ connection 208 is preferably added for structural integrity of the EL panel.

[0048] The lightbox assembly further comprises a power source 210 and a custom-designed printed circuit board (PCB) 212. The power source 210 may be provided in the form of a lightweight, rechargeable battery, such as a lithium-polymer (LiPo) battery. The custom PCB 212 is configured to manage power functions, including charging and voltage regulation of the power source 210. The access to both PCB 212 and power source 210 is protected by the lid 214.

[0049] The lightbox assembly 200 creates a controlled lighting environment, which was found to be very useful for consistent and accurate analysis of the baked goods crumb structure. In some embodiments of the invention, the method includes using a flash to control a light / shadow distribution on a surface of the baked goods item. In alternative embodiments the base of the lightbox is designed with a pre-defined color such as white, to ensure accurate white balance is achieved during the acquisition of visual data.

[0050] Figure 3 shows an example of the light distribution in the lightbox 300. The EL light panel 302 emits light parallel to the surface of the baked goods item 304, which is positioned inside the lightbox, providing illumination from each side and across its full height. This arrangement ensures that shadows are cast only in the structural voids of the baked goods item, accurately reflecting its crumb structure. For example, image 352 demonstrates the difference in light distribution between conventional industry systems 354 and embodiments of the present invention 356. In conventional systems, visible marks from crust deformation 358 and slicing knives 360 on the bread slice are misidentified as voids due to suboptimal light distribution. However, when the same baked goods item is analysed using an embodiment of the present invention, no shadows are formed around the crust deformation 362 and uneven slicing 364 of the baked goods item, resulting in improved accuracy in detecting crumb structure by preventing false void detection. This enhanced light distribution provides more reliable data regarding the crumb structure properties of the baked goods item.

[0051] The above description has resulted in the collection of the visual data regarding the baked goods item using the smartphone. This collected information is then fed into a series of software modules, as shown in Figure 4.

[0052] In Figure 4, in accordance with the preferred embodiment of the invention, the system for analysing a baked goods item, is generally described with reference to numeral 400.

[0053] The system 400 is provided in the example form of a mobile device 40 which includes the camera 402, the LiDAR sensor 404 and software 406. The system 400 further includes a database 408.

[0054] In use, the software 406 (which is comprised of a first software module 410, a second software module 412 and a third software module 414, described below) receives the visual data from the camera 402 and LiDAR sensor 404 and processes such data to perform an analysis of the crumb structure of the baked goods item as will be described below.

[0055] The first software module 410 carries out segmentation of the received visual data from the camera 402 to identify the contour of the baked goods item 106. This means that the position of the baked goods item 106 on the overall image(s) or video(s) is identified, and the position of the baked goods item is separated or “segmented” from the rest of the image(s) or video(s) content. In an embodiment, the step of segmenting the visual data includes use of a salient object detection (SOD) methodology. In this embodiment, the salient object detection methodology is provided in the example form of an Artificial Intelligence (Al) tool which is operable to extract a contour of the analysed baked goods item. In this embodiment, the Al tool is provided in the example form of a U-2 Net model. It is to be appreciated that different neural networks could also be used for image segmentation.

[0056] The identified contour of the baked goods item is further sent to the second software module 412 where another stage of segmentation of the internal area of the contour is performed. In an embodiment of the invention, the step of segmenting the area inside the contour of the baked goods item involves dividing an image into multiple segments or regions based on similar characteristics. In an embodiment, image segmentation is completed by a thresholding technique. The threshold in the thresholding technique is defined as a specific intensity value that serves as a boundary to classify pixels within the image. The threshold is selected to minimise the variance within the intensity groups and maximise the variance between them. Once the threshold is applied, the image is converted into a two-tone representation (also known as binary representation) as shown in Figure 6. Pixels with intensity values below the threshold correspond to voids or open spaces such as pores or holes 602, while pixels with higher intensity values represent solid or filled areas, such as the crumb 604. This binary transformation allows for the clear identification and separation of regions.

[0057] In another embodiment, the step of segmenting the area within the contour includes applying a clustering algorithm to the image, for example a k-means clustering algorithm. The k- means clustering algorithm is a well-known method of vector quantization, that partitions (n) observations into (k) clusters in which each observation belongs to the cluster with the nearest mean, serving as a prototype of the cluster. The clustering algorithm groups a set of data into clusters where data arrays consist of pixels of different intensity. Each cluster formed by the k-means clustering algorithm may comprise multiple data arrays, depending on the structure and segmentation of the image data. A single data array can include pixels of varying intensities, as the clustering algorithm evaluates and assigns pixels to clusters based on intensity similarity relative to a centroid. Within this context, the centroid represents the average value of the features (e.g., pixel intensities) of the data points within a cluster. This allows for the grouping of pixels into clusters, facilitating the segmentation of distinct regions within the image. For example, the k-means algorithm with k=2 produces a binary file similar to that shown in Figure 6.

[0058] In some embodiments, a step of image pre-processing could be used to enhance the quality of the image, reduces noise, improves contrast, and prepares the image for analysis and further processing. This step of pre-processing the image could include one or more of the following techniques: colour normalisation, contrast pre-processing, and / or noise reduction. The step of image pre-processing could be used for image standardisation.

[0059] The final segmented data, in combination with the distance data from the smartphone 40 to the baked goods item, as received by the LiDAR sensor 404, is forwarded to the third software module 414 for estimating a plurality of crumb structure characteristics, including but not limited to porosity, cell wall size, and cell diameter. In a current embodiment, the calculation of the spatial resolution, described above, is performed in the third module 414. In yet another embodiment the calculation of the spatial resolution can be performed at any stage earlier and stored until it is required for estimation of crumb structure characteristics. Estimation of a plurality of the crumb structure characteristics are performed utilising known image processing methodologies, including but not limited to, those based on image processing libraries such as OpenCV. The output of the third software module 414 is the plurality of crumb structure properties, such as porosity level, cell diameter, cell wall size, concavity level etc. and is further illustrated in Figure 7. In operation, database 408 houses the product quality parameters, which can be compared to the results generated from the software 406, to provide a product quality assessment result.

[0060] Because of the use of the lightbox during obtaining of the visual data, software module 412, as described in Figure 3, illustration 352, receives high-quality visual data regarding the light-shadow distribution on the surface of the baked goods item 106. Simultaneously, the low near-infrared radiation emitted by the EL panel 302, used as the light source, does not interfere with the LiDAR pulses 108a-b, thereby allowing for accurate acquisition of visual data pertaining to the distance between the baked goods item 106 and the smartphone 102, as shown in Figure 1. In the current embodiment, all software modules are executed on the smartphone. However, in alternative embodiments, the modules may be executed either on the cloud or on the smartphone, either individually, collectively, or in any combination thereof.

[0061] Figure 5a shows the smartphone device 500 and the view of the open mobile application screen 502 before capturing the visual data of the baked goods item 504. The visible line frame 506 on the screen, guides user to adjust the distance of the smartphone 500 to the baked goods item 504 so it visually fits the frame 506. A capture button 508 is provided, enabling the user to initiate the acquisition of visual data of the baked goods item and to identify contour of the baked goods item.

[0062] Figure 5b shows smartphone device 550 with the open mobile application screen 552 where identified contour 554 of the baked goods item is illustrated together with its internal area. A confirmation button 556 is provided, enabling the user to validate the accuracy of the contour segmentation and submit the data about the contour and its internal area for further processing. In case the contour segmentation must be re-taken, the user has the possibility to do so by clicking a button 558, that will bring the user to the previous process step as described in Figure 5a. The received data about real-time distance 560 between the smartphone and the baked goods item is displayed.

[0063] Figure 6 illustrates an image of the baked goods item 600 after the user activates the confirmation button 556. This is an example of a binary representation of the image after it is processed by software module 412 where the pixels with different intensities are grouped together. As a result, the black regions 602 in the image represent voids, such as pores or holes, while the white regions 604 represent the crumb. This image is sent further to software module 414 for estimation of the plurality of crumb structure characteristics and show analysis results to the user as described below.

[0064] Figure 7 shows a smartphone device 700 with the mobile application screen 702 displayed, where the results of the crumb structure analysis of the baked goods items 704a-f are shown. The crumb structure properties of each analysed baked goods item 704a-f are estimated and displayed. For example, porosity 706 is shown for slice 704d. Additionally, the concavity level 708 and the cell wall size 710 of the baked goods items 704d and 704e respectively are presented. In an embodiment, the image is processed in real-time. In an alternative embodiment, the image is queued for later analysis.

[0065] In Figure 8, a flow chart showing a method of analysing a baked goods item, in accordance with an aspect of the invention, is generally described with reference to numeral 800.

[0066] The method includes, at block 802, a first step of opening the mobile application and at block 804 a second step of creating a project (e.g. folder) to store the visual data and results of crumb structure characteristics estimation. This step can be done before or after the analysis of the item. In this embodiment the project is stored on a cloud storage. In yet another embodiment the project can be stored on the smartphone.

[0067] The next step at block 806 includes placing the baked goods item inside the lightbox. The action at the next step of block 808 involves enabling the camera and positioning it above the baked goods item. It will be appreciated, that a 180° angle between surface of the analysed product and mobile device is ideal. Within this context, the camera on the mobile device provides the focus on the object with / without LiDAR sensor support.

[0068] Once the analysed area is in focus; the visual data can be taken at block 810. There can be 0-30s delay from receiving the visual data, for the LiDAR sensor to identify the distance to the baked goods item. In this embodiment, the operation of receiving visual data is preferably performed using both the RGB camera(s) (810a) and the LiDAR sensor (810b). Alternatively, the focus point between 2 or more cameras can be used for distance measurement or a combination of the distance data provided by the LiDAR sensor and the data from the 2 or more cameras (i.e., stereo cameras) can be used.

[0069] At block 812 a U2-Net Model (812a) is utilised to segment the received visual data and identify the contour of the analysed baked good item.

[0070] At block 814, the step of segmenting the area inside the identified contour (as described in module 414) involves dividing an image into multiple segments or regions based on similar characteristics.

[0071] Process at block 816 enables further analysis of the baked goods item by allowing the estimation of the plurality of crumb structure characteristics. By clearly distinguishing between the crumb and the voids or open spaces (i.e., pores, holes), various structural properties can be quantified, including the characteristics that requires spatial resolution value. In current embodiment, the spatial resolution is calculated in the block 816a.

[0072] At block 818 visual data and analysis results are saved into the database.

[0073] In a current embodiment, the analysis results are further shown to the user at block 820 as also illustrated in Figure 7.

[0074] In Figure 9, a computer within which a set of instructions, for causing the computer to perform any one or more of the methodologies described herein, may be executed. In accordance with embodiments of the invention, the computer is generally described with reference to numeral 900.

[0075] In a networked deployment, the computer 900 may operate in the capacity of a server or a client machine in server-client network environment, or as a peer machine in a peer-to-peer (or distributed) network environment. The computer 900 may be a personal computer (PC), a tablet PC, a set-top box (STB), a Personal Digital Assistant (PDA), a cellular telephone, a web appliance, a network router, switch or bridge, or any computer 900 capable of executing a set of instructions (sequential or otherwise) that specify actions to be taken by that computer 900.

[0076] Further, while only a single computer 900 is illustrated, the term "computer" shall also be taken to include any collection of computers that individually or jointly execute a set (or multiple sets) of instructions to perform any one or more of the methodologies discussed herein.

[0077] The example computer system 900 includes a processor 902 (e.g., a central processing unit (CPU), a graphics processing unit (GPU) or both), a main memory 904 and a static memory 906, which communicate with each other via a bus 901. The computer 900 may further include a video display unit 910 (e.g., a liquid crystal display (LCD)). The computer 900 also includes an alphanumeric input device 912 (e.g., a keyboard), a user interface (Ul) navigation device 914 (e.g., a mouse), a disk drive unit 916, a signal generation device 918 (e.g., a speaker) and a network interface device 908.

[0078] The disk drive unit 916 includes a computer-readable medium 922 on which is stored one or more sets of instructions and data structures (e.g., software 924) embodying or utilizing any one or more of the methodologies or functions described herein. The software 924 may also reside, completely or at least partially, within the main memory 904 and / or within the processor during execution thereof by the computer system 900, the main memory and the processor also constituting computer-readable media. To this end, for clarity, please note that where the software 924 is not located in the main memory 904 and / or within the processor during execution thereof by the computer system 900, it will be located in a cloudbased or remote storage location and may be executed directly from there.

[0079] The software 924 may further be transmitted or received over a network 926 via the network interface device 908 utilizing any one of several well-known transfer protocols (e.g., secure HTTP, FTP). While the computer-readable medium 922 is shown in an example embodiment to be a single medium, the term "computer-readable medium" should be taken to include a single medium or multiple media (e.g., a centralized or distributed database, and / or associated caches and servers) that store the one or more sets of instructions.

[0080] The term "computer-readable medium" shall also be taken to include any medium that is capable of storing, encoding or carrying a set of instructions for execution by the computer 900 and that cause the computer 900 to perform any one or more of the methodologies of the present embodiments, or that is capable of storing, encoding or carrying data structures utilised by or associated with such a set of instructions.

[0081] The term "computer-readable medium" shall accordingly be taken to include, but not be limited to, solid-state memories and optical and magnetic media as well as cloud storage options (such as Amazon Webservices ™, Microsoft Azure ™, and the like). It is to be understood that the invention is not limited to the specific details described herein which are given by way of example only and that various modifications and alterations are possible without departing from the scope of the invention as defined in the appended claims.

Claims

CLAIMS:

1. A method of analysing a crumb structure of a baked goods item, comprising steps of: a) receiving at a mobile device, visual data of the baked goods item, including realtime distance measurements of the distance from the mobile device to the baked goods item, wherein the visual data has been collected in a light controlled environment; b) segmenting the received visual data to identify a contour of the baked goods item; c) segmenting the area inside the contour of the baked goods item; and d) estimating a plurality of crumb structure properties of the baked goods item, using the results of step c), by separating regions within the baked goods item based on pixel intensity wherein regions of lower intensity correspond to open spaces or voids, and regions with higher intensity correspond to the crumb.

2. The method as claimed in claim 1, wherein the plurality of crumb properties include the porosity level, cell wall size, and concavity level of the baked goods item.

3. The method as claimed in any of the preceding claims, wherein the step of receiving visual data of the baked goods item includes the step of accessing a camera and a LiDAR sensor, on the mobile device, to capture image(s) of the item and / or determine a distance to the item.

4. The method as claimed in claim 3, wherein the step of receiving visual data includes enabling a camera on the mobile device and positioning it above the item, in a lateral position to the object.

5. The method as claimed in claim 1 , wherein the step d) is carried out using a calculated spatial resolution.

6. The method as claimed in claim 5, wherein the spatial resolution is calculated from the sensor’s true height size and the image resolution.

7. The method as claimed in any of the preceding claims, wherein the method includes using a flash to control a light / shadow distribution on the surface of the baked goods item.

8. The method as claimed in any of claims 7, wherein the method includes using one or more application algorithms to calculate flash mode parameters for use in the real time processing of the image.

9. The method as claimed in any of the preceding claims, wherein the step (b) includes use of a salient object detection (SOD) methodology.

10. The method as claimed in claim 9, wherein the salient object detection methodology is provided in the example form of an Artificial Intelligence (Al) tool which is operable to extract one or more borders of the analysed baked goods item.

11. The method as claimed in claim 10, wherein the Al tool is provided in the example form of a U-2 Net model.

12. The method as claimed in claim 1, wherein the step (c) applies a threshold technique where a threshold is selected, where the threshold is defined as a specific intensity value that serves as a boundary to classify pixels within the image.

13. The method as claimed in claim 12, wherein after the threshold technique is applied, the image is converted into a two-tone representation wherein a segmentation algorithm groups pixels according to their intensity levels related to the selected threshold.

14. The method of claim 1, wherein the step (c) uses a k-means clustering algorithm.

15. The method as claimed in claim 14, wherein the k-means clustering algorithm is a method of vector quantization, that partitions (n) observations into (k) clusters in which each observation belongs to the cluster with the nearest mean.

16. The method as claimed in any of claims 14 or 15, wherein the clustering algorithm groups a set of data into a plurality of clusters, wherein each cluster has a plurality of data arrays, wherein each data array includes pixels representing various intensities.

17. The method as claimed in claim 16 wherein the clustering algorithm assigns pixels to clusters based on instensity value similarity relatative to a centroid.

18. The method as claimed in claim 17 wherein the centroid represents the average valaue of pixel intensities of data points within a cluster.

19. The method as claimed in claim 1, wherein the light controlled environment includes a light source configured to emit minimal or no radiation in the near-infrared range.

20. The method as claimed in claim 19 and the light source is configured to emit light parallel to the surface of the baked goods item wherein the baked goods item is located in a lightbox, wherein the light source is located inside the lightbox in sucha configuration that the light source emits light from each of a plurality of sides of the baked goods item and across the height of the baked goods item.

21. A system comprising means adapted for carrying out all the steps of the method according to any preceding method claim.

22. A computer program comprising instructions for carrying out all the steps of the method according to any preceding method claim, when said computer program is executed on a computer system.

Images