Determining object boundaries

Abstract

A method and system for determining object boundaries. Time-of-flight signals for a plurality of regions of an environment containing a first object and a second object are received and processed to identify at least one boundary region containing the first and second objects, at least one first reference region containing only the first object and at least one second reference region containing only the second object. The relative signal strengths of the ToF signals for a boundary region, first reference region and second reference region are used to estimate a proportion of the boundary region occupied by each of the first and second objects, and the estimated proportions are processed to determine a boundary position within the boundary region.

Description

[0001] DETERMINING OBJECT BOUNDARIES

[0002] FIELD OF THE INVENTION

[0003] The invention relates to the field of determining boundaries of objects.

[0004] BACKGROUND OF THE INVENTION

[0005] Many sensor-based smart lighting systems control the provision of light based on a person’s movements with respect to a particular region of interest within an environment. The region of interest may be defined with reference to a specific object in the environment. For instance, a smart lighting system may be configured to control a lamp on or next to a desk to provide light when a person moves within a predefined distance of the desk.

[0006] In many smart lighting systems, a low-resolution (e.g. 64 pixels) time-of-flight sensor is used to determine a location of an object that defines the region of interest and to detect a person’s movements with respect to the region of interest. The use of a low- resolution time-of-flight sensor enables a user of a smart lighting system to preserve privacy; on the other hand, the low resolution reduces the precision of the determination of the location of the object that defines the region of interest (and, in particular, the determination of the boundary of the object).

[0007] There is therefore a need for an improved method for determining object boundaries.

[0008] SUMMARY OF THE INVENTION

[0009] The invention is defined by the claims.

[0010] According to examples in accordance with an aspect of the invention, there is provided a computer-implemented method for determining object boundaries, the computer- implemented method comprising: receiving, from a time-of-flight, ToF, sensor, a ToF signal for each of a plurality of regions of an environment; processing the ToF signals to identify a boundary region, wherein the boundary region is a region of the environment containing a boundary between a first object, located at a first distance from the ToF sensor, and a second object, located at a second, different distance from the ToF sensor; processing the ToF signals to identify a first reference region, wherein the first reference region is a region of the environment that is predicted to contain only the first object; processing the ToF signals to identify a second reference region, wherein the second reference region is a region of the environment that is predicted to contain only the second object; estimating a proportion of the boundary region occupied by the first object based on a ratio of a signal strength of the ToF signal for the boundary region for the first distance to a signal strength of the ToF signal for the first reference region; estimating a proportion of the boundary region occupied by the second object based on a ratio of a signal strength of the ToF signal for the boundary region for the second distance to a signal strength of the ToF signal for the second reference region; and processing the proportion of the boundary region occupied by the first object and the proportion of the boundary region occupied by the second object to determine a boundary position for the boundary between the first object and the second object within the boundary region.

[0011] This allows the boundary between two objects to be determined to a greater precision than is otherwise possible for a given spatial resolution of the time-of-flight sensor. The inventors have recognized that the signal strength of a ToF signal for an object depends on the proportion of the region (from which the ToF signal was acquired) occupied by the object, and that, therefore, a position of a boundary of the object within a boundary region may be determined by comparing the signal strength for the object between a region fully occupied by the object and the boundary region.

[0012] For instance, if the first object and the second object occupy equal proportions of the boundary region, a boundary position that divides the boundary region in half may be determined as the boundary region.

[0013] In some examples, each ToF signal is acquired by a single pixel of the ToF sensor. In other words, each of the plurality of regions is a region covered by a single pixel of the ToF sensor. A pixel of a ToF sensor is a sensor element that defines the spatial resolution of the ToF sensor; in other words, light reflected from one point and detected by the ToF sensor cannot be spatially differentiated from light reflected from another point having a same distance from the ToF sensor that is detected by the same pixel of the ToF sensor.

[0014] In some examples, the first reference region is a region of the environment for which the ToF signal has a single peak corresponding to the first distance; and the second reference region is a region of the environment for which the ToF signal has a single peak corresponding to the second distance. In some examples, the first reference region and / or the second reference region are adjacent to the boundary region. This may improve an accuracy of the determined boundary.

[0015] In some examples, the method may further comprise, in response to a determination that a sum of the proportion of the boundary region occupied by the first object and the proportion of the boundary region occupied by the second object, when expressed as a percentage, is outside a predefined percentage range: processing the ToF signals to identify a new first reference region; estimating a new proportion of the boundary region occupied by the first object based on a ratio of a signal strength of the ToF signal for the boundary region for the first distance to a signal strength of the ToF signal for the new first reference region; and processing the new proportion of the boundary region occupied by the first object to determine the boundary position within the boundary region.

[0016] In some cases, a region identified as a first reference region may not be entirely occupied by the first object. For instance, a region occupied by an object having approximately the same distance to the ToF sensor as the first object might be identified as a first reference region. If the reflective properties of this object are sufficiently different to the reflective properties of the first object, this would result in an inaccurate estimate of the proportion of the boundary region occupied by the first object. Such a situation may be identified by comparing the sum of the determined proportions to a predefined percentage range, and the first reference region may be re-identified if the sum falls outside the predefined percentage range.

[0017] In some examples, the method may further comprise, in response to a determination that a sum of the proportion of the boundary region occupied by the first object and the proportion of the boundary region occupied by the second object, when expressed as a percentage, is outside the predefined percentage range: processing the ToF signals to identify a new second reference region; estimating a new proportion of the boundary region occupied by the second object based on a ratio of a signal strength of the ToF signal for the boundary region for the second distance to a signal strength of the ToF signal for the new second reference region; and processing the new proportion of the boundary region occupied by the second object to determine the boundary position within the boundary region.

[0018] In some cases, a region identified as a second reference region may not be entirely occupied by the second object, resulting in an inaccurate estimate of the proportion of the boundary object occupied by the second object. In some examples, the predefined percentage range has a lower limit between 80% and 90%, and an upper limit between 110% and 120%.

[0019] In some examples, the method may further comprise: processing the ToF signals to identify one or more further boundary regions, wherein each further boundary region is a region of the environment containing a boundary between the first object and the second object; and for each further boundary region: estimating a proportion of the further boundary region occupied by the first object based on a ratio of a signal strength of the ToF signal for the further boundary region for the first distance to a signal strength of the ToF signal for the first reference region; estimating a proportion of the further boundary region occupied by the second object based on a ratio of a signal strength of the ToF signal for the further boundary region for the second distance to a signal strength of the ToF signal for the second reference region; processing the proportion of the further boundary region occupied by the first object and the proportion of the further boundary region occupied by the second object to determine a boundary position for the boundary between the first object and the second object within the further boundary region.

[0020] In some examples, the determination of each boundary position is further based on the proportion occupied by the first object and the proportion occupied by the second object for one or more other boundary regions. For instance, the difference in proportions between adjacent boundary regions may be used to infer the angle of the boundary.

[0021] In some examples, the method may further comprise processing each determined boundary position to generate a layout of the environment. The layout of an environment may, for example, be used in smart lighting systems or other smart appliances, home security systems, and healthcare monitoring systems.

[0022] In some examples, the method may further comprise: processing each determined boundary position to define a detection zone in the environment; processing the ToF signal for each region of the environment in the detection zone to detect, if present, a movement in the detection zone; and controlling a lighting device to provide light in response to detecting a movement in the detection zone.

[0023] For instance, a desk lamp may be controlled to provide light in response to a person moving within a predetermined distance of a desk. In another example, a night light may be controlled to provide light in response to a person leaving their bed. There is also proposed a computer program product comprising computer code means which, when executed on a computing device having a processing system, cause the processing system to perform all of the steps of any of the methods described above.

[0024] According to examples in accordance with another aspect of the invention, there is provided a processing system for determining object boundaries, the processing system being configured to: receive, from a time-of-flight, ToF, sensor, a ToF signal for each of a plurality of regions of an environment; process the ToF signals to identify a boundary region, wherein the boundary region is a region of the environment containing a boundary between a first object, located at a first distance from the ToF sensor, and a second object, located at a second, different distance from the ToF sensor; process the ToF signals to identify a first reference region, wherein the first reference region is a region of the environment that is predicted to contain only the first object; process the ToF signals to identify a second reference region, wherein the second reference region is a region of the environment that is predicted to contain only the second object; estimate a proportion of the boundary region occupied by the first object based on a ratio of a signal strength of the ToF signal for the boundary region for the first distance to a signal strength of the ToF signal for the first reference region; estimate a proportion of the boundary region occupied by the second object based on a ratio of a signal strength of the ToF signal for the boundary region for the second distance to a signal strength of the ToF signal for the second reference region; and process the proportion of the boundary region occupied by the first object and the proportion of the boundary region occupied by the second object to determine a boundary position for the boundary between the first object and the second object within the boundary region.

[0025] There is also proposed a system for determining object boundaries, the system comprising a time-of-flight, ToF, sensor and the processing system described above.

[0026] In some examples, the system further comprises a lighting device, wherein the processing system is configured to control the lighting device based on the determined boundary position.

[0027] These and other aspects of the invention will be apparent from and elucidated with reference to the embodiment s) described hereinafter.

[0028] BRIEF DESCRIPTION OF THE DRAWINGS

[0029] For a better understanding of the invention, and to show more clearly how it may be carried into effect, reference will now be made, by way of example only, to the accompanying drawings, in which:

[0030] Fig. 1 illustrates a system for determining object boundaries, according to an embodiment of the invention;

[0031] Fig. 2 illustrates the field of view of a ToF sensor; and

[0032] Fig. 3 illustrates a computer-implemented method for determining object boundaries, according to an embodiment of the invention.

[0033] DETAILED DESCRIPTION OF THE EMBODIMENTS

[0034] The invention will be described with reference to the Figures.

[0035] It should be understood that the detailed description and specific examples, while indicating exemplary embodiments of the apparatus, systems and methods, are intended for purposes of illustration only and are not intended to limit the scope of the invention. These and other features, aspects, and advantages of the apparatus, systems and methods of the present invention will become better understood from the following description, appended claims, and accompanying drawings. It should be understood that the Figures are merely schematic and are not drawn to scale. It should also be understood that the same reference numerals are used throughout the Figures to indicate the same or similar parts.

[0036] The invention provides a method and system for determining object boundaries. Time-of-flight signals for a plurality of regions of an environment containing a first object and a second object are received and processed to identify at least one boundary region containing the first and second objects, at least one first reference region containing only the first object and at least one second reference region containing only the second object. The relative signal strengths of the ToF signals for a boundary region, first reference region and second reference region are used to estimate a proportion of the boundary region occupied by each of the first and second objects, and the estimated proportions are processed to determine a boundary position within the boundary region.

[0037] Embodiments are at least partly based on the realization that the signal strength of a ToF signal for a given distance within a region depends on the proportion of the region occupied (from the viewpoint of the ToF sensor acquiring the ToF signal) by an object at that distance.

[0038] Illustrative embodiments may, for example, be employed in smart lighting systems, smart appliances, security systems and healthcare monitoring systems. Figure 1 illustrates a system 100 for determining object boundaries, according to an embodiment of the invention. The system comprises a time-of-flight, ToF, sensor 110 and a processing system 120. The processing system 120 is, itself, an embodiment of the invention. In Figure 1, the system 100 is provided in an environment having a first object, a desk 130, and a second object, a floor 140, which are located at different distances from the ToF sensor. In some examples, the system may further comprise a lighting device 150.

[0039] The ToF sensor 110 may be any sensor configured to emit and detect light 111 (e.g. infrared radiation), and output, for each of a plurality of regions, a signal responsive to a time of flight of light emitted by the ToF sensor and reflected back to the ToF sensor. For instance, the ToF sensor may emit a short pulse of light, and measure the time taken to detect a reflection of the pulse of light. Alternatively, the ToF sensor may emit a continuous wave of modulated light; the phase of the detected light would then depend on the time of flight of the light.

[0040] In some examples, the ToF sensor is a low-resolution ToF sensor. Whether a ToF sensor is considered to be a low-resolution sensor may depend on an application of the system 100 and the distance to the regions for which a ToF signal is acquired. For instance, a ToF sensor may be considered to be a low-resolution ToF sensor if the size of the region corresponding to each pixel of the ToF sensor exceeds a desired precision for boundary determination. For example, when used in a smart lighting system, the ToF sensor may be considered to have a low resolution if each pixel corresponds to a region of at least 10 cm * 10 cm.

[0041] The ToF sensor 110 is configured to acquire a time-of-flight, ToF, signal 115 for each of a plurality of regions of an environment. Each ToF signal is a signal responsive to the time of flight of light emitted by the ToF sensor and reflected by a respective region of the environment. For instance, each ToF signal may be a signal directly recording the time of flight for light reflected by the respective region or a signal recording the distance travelled by the light. Conceptually, each of the plurality of regions may be considered as a region of a projection plane of the ToF sensor.

[0042] For example, in Figure 1, the ToF sensor 110 is positioned such that the light 111 emitted by the ToF sensor is incident on and reflected by the desk 130 and the floor 140. The plurality of regions for which ToF signals are acquired by the ToF sensor therefore includes regions containing only the desk (from the perspective of the ToF sensor), regions containing only the floor, and regions containing both the desk and the floor. Each of the plurality of regions corresponds to at least one pixel of the ToF sensor 110. Preferably, each region corresponds to a same number of pixels; if the regions have different sizes (in terms of the number of pixels covering the region), the ToF signals may be adjusted according to the size of the region before the signals are used to determine an object boundary. In some examples, each region may correspond to a single pixel of the ToF sensor, i.e. each region is defined by the field of view of a pixel of the ToF sensor. In other words, each ToF signal 115 may be acquired by a different single pixel of the ToF sensor.

[0043] The processing system 120 is configured to receive the ToF signals 115 for the plurality of regions from the ToF sensor 110. The processing system is then configured to process the ToF signals to identify a boundary region, a first reference region and a second reference region.

[0044] A boundary region is a region of the plurality of regions containing a boundary between a first object (in Figure 1, the desk 130), located at a first distance from the ToF sensor 110, and a second object (in Figure 1, the floor 140), located at a second, different distance from the ToF sensor. A region for which the ToF signal has two peaks, corresponding to different distances, may be identified as a boundary region. The distance corresponding to one peak may then be identified as the first distance, and the distance corresponding to the other peak may be identified as the second distance.

[0045] The first reference region is a region of the plurality of regions that is predicted to contain only the first object. A region for which the ToF signal has a single peak that corresponds to the first distance (i.e. a single peak at the first distance value or at a time value corresponding to the first distance) may be identified as the first reference region.

[0046] The second reference region is a region of the plurality of regions that is predicted to contain only the second object. A region for which the ToF signal has a single peak that corresponds to the second distance (i.e. a single peak at the second distance value or at a time value corresponding to the second distance) may be identified as the second reference region.

[0047] As the skilled person will readily appreciate, different areas of the same surface of an object may have slightly different distances to the ToF sensor. A region may therefore be identified as containing a peak corresponding to the first or second distance if the distance is within a predefined percentage of the first or second distance respectively. The predefined percentage may depend on the value of the first / second distance and / or on the distance between the region and the boundary region. In some examples, the first reference region may be identified as a region predicted to contain the first object that is adjacent to or closest to the identified boundary region. Similarly, the second reference region may be identified as a region predicted to contain the second object that is adjacent to or closest to the identified boundary region. This may improve a likelihood that the first and second reference regions contain the first and second objects respectively (i.e. rather than another object having the same distance from the ToF sensor), and increase an accuracy of a boundary position determined based on the reference regions.

[0048] The identification of the boundary region and first and second reference regions may be described more clearly with reference to Figure 2, which illustrates the field of view 200 of a ToF sensor 210 having 64 pixels. The dotted lines divide the field of view into a plurality of regions, each corresponding to a field of view of a pixel of the ToF sensor.

[0049] In Figure 2, the ToF sensor 210 is positioned at a height of 190 cm from a floor 240, above a desk 230 having a height of 80 cm from the floor. This means that a region of the desk covered by a single pixel of the ToF sensor has an area of 20 cm x 20 cm, and a region of the floor covered by a single pixel of the ToF sensor has an area of 34 cm x 34 cm. The white area of the field of view 200 represents the transition between the desk and floor at the boundary between the desk and floor. Two additional objects are provided on the desk: a box 260 and a sheet of paper 270.

[0050] The ToF sensor 210 is configured to acquire a ToF signal at each of pixels 1- 64. Pixels 9, 16, 17, 24, 25, 32, 33-37 detect light reflected by both the desk and the floor. The ToF signals acquired by each of these pixels would include peaks corresponding to two distinct distances (approximately 110 cm and 190 cm); the regions corresponding to these pixels would therefore be identified as boundary regions. All light detected by pixels 12-15, 20-23 and 26-29 has been reflected by the desk 230, so the ToF signals acquired by these pixels would contain a single peak corresponding to a distance of approximately 110 cm, and the regions corresponding to these pixels could be identified as first reference regions. All light detected by pixels 1-8 and 41-64 has been reflected by the floor 240, so the ToF signals acquired by these pixels would contain a single peak corresponding to a distance of approximately 190 cm, and the regions corresponding to these pixels could be identified as second reference regions.

[0051] Pixels 10, 11, 18 and 19 detect light reflected by the desk 230 and the box 260. The box 260 has a height large enough to enable the box to be distinguished from the desk 230 by the ToF sensor; the ToF signals acquired by these pixels would be identified as unsuitable for the first reference region, because, as well as a peak corresponding to a distance of approximately 110 cm, these signals would include an additional peak corresponding to a different distance.

[0052] Pixels 30 and 31 detect light reflected by the desk 230 and the sheet of paper 270. As a sheet of paper is relatively thin, the ToF signals acquired by these pixels would be identified as having a single peak corresponding to a distance of approximately 110 cm, and the regions corresponding to these pixels may be identified as first reference regions. Similarly, pixels 38 and 39 detect light reflected by the desk, the sheet of paper and the floor 240, so the ToF signals acquired by these pixels would be identified as having a first peak corresponding to a distance of approximately 110 cm and a second peak corresponding to a distance of approximately 190 cm. The regions corresponding to pixels 38 and 39 would therefore be identified as boundary regions.

[0053] Returning to Figure 1, having identified a boundary region and a first reference region, the processing system 120 is configured to estimate a proportion of the boundary region that is occupied by the first object, based on a ratio of a signal strength of the ToF signal for the boundary region for the first distance to a signal strength of the ToF signal for the first reference region.

[0054] The processing system 120 is configured to estimate a proportion of the boundary region occupied by the second object based on a ratio of a signal strength of the ToF signal for the boundary region for the second distance to a signal strength of the ToF signal for the second reference region.

[0055] In some examples, the signal strengths used to estimate the proportions of the boundary region occupied by the first and second objects may be average signal strengths, obtained by averaging repeat measurements of the relevant signal strength.

[0056] The processing system 120 is then configured to process the proportion of the boundary region occupied by the first object and the proportion of the boundary region occupied by the second object to determine a boundary position for the boundary between the first object and the second object.

[0057] For instance, if the proportions of the boundary region occupied by the first object and second object respectively, when expressed as percentages, sum to exactly 100%, the processing system may determine the boundary position as a position that divides the boundary region into a first subregion, occupying a proportion of the boundary region equal to the proportion of the boundary region occupied by the first object, and a second subregion, occupying a proportion of the boundary region equal to the proportion of the boundary region occupied by the second object. For example, if the first object and second object are each determined to occupy 50% of the boundary region, the determined boundary position may be a position that divides the boundary region exactly in half.

[0058] If the proportions of the boundary region occupied by the first object and second object respectively, when expressed as percentages, do not sum to exactly 100%, the processing system may determine the proportion occupied by the first object as a first percentage of the total and the proportion occupied by the second object as a second percentage of the total. The processing system may then determine the boundary position as a position that divides the boundary region into a first subregion and a second subregion, where the first subregion occupies a proportion of the boundary region equal to the first percentage and the second subregion occupies a proportion of the boundary region equal to the second percentage. For instance, if the first object is determined to occupy 40% of the boundary region and the second object 70% of the boundary region, the determined boundary position may be a position that divides the boundary region into a first subregion occupying 36% of the boundary region and a second subregion occupying 64% of the boundary region.

[0059] Tables I and II demonstrate how this method enables the determination of a boundary position within a boundary region. Table I provides distance measurements and IR signal strength acquired by an infrared ToF sensor for boundary regions containing a desk and a floor, and a first reference region containing only the desk. Table II provides distance measurements and IR signal strength acquired by an infrared ToF sensor for boundary regions containing the desk and the floor, and a second reference region containing only the floor, and the sum of the estimated proportion occupied by the desk and the estimated proportion occupied by the floor. In each Table, Nos. 1-20 represent repeat measurements for each boundary region and the reference region.

[0060] TABLEI TABLE II The results provided in Tables I and II show that a relatively accurate estimate of the proportions of each boundary region occupied by the desk and floor may be obtained using this method. For instance, the 25% desk boundary region is estimated to have 31% occupied by the desk and 80% occupied by the floor, summing to 111%. The determined boundary position for this boundary region may therefore be a position that divides the boundary region into a first subregion occupying 28% (31 111 x 100) of the region and a second subregion occupying 72% (80 ^- 111 * 100) of the region. Similarly, the 50% desk boundary region is estimated to have 58% occupied by the desk and 52% occupied by the floor, summing to 110%. The determined boundary position for this boundary region may therefore be a position that divides the boundary region into a first subregion occupying 53% (58 110 x 100) of the region and a second subregion occupying 47% (52 ^- 110 x 100) of the region. The 75% desk boundary region is estimated to have 85% occupied by the desk and 23% occupied by the floor, summing to 108%. The determined boundary position for this boundary region may therefore be a position that divides the boundary region into a first subregion occupying 79% (85 108 x 100) of the region and a second subregion occupying 21% (23 108 x l00) of the region. Each of these determined boundary positions thus divides the respective boundary region into subregions occupying proportions that are within ±5% of the actual proportions occupied by the desk and floor.

[0061] In some examples, the sum of the proportions of the boundary region occupied by the first object and second object respectively may be compared to a predefined percentage range before determining the position of the boundary range. If the boundary region contains only the first and second objects, the first reference region only the first object and the second reference region only the second object, it is to be expected that the sum of the proportions will, when expressed as a percentage, be close to 100%. The predefined percentage range may therefore be a range including 100%. In some examples, the predefined percentage range may have a lower limit between 80% and 90%, and an upper limit between 110% and 120%. For instance, the predefined percentage range may be a range of 80-110%, 80-115%, 80-120%, 85-110%, 85-115%, 85-120%, 90-110%, 90-115%, or 90-120%.

[0062] If the sum of the proportion of the boundary region occupied by the first object and the proportion of the boundary region occupied by the second object, when expressed as a percentage, falls within the predefined percentage range, it may be determined that the identified boundary region, first reference region and second reference region are suitable for determining a position of a boundary between the first object and the second object, and the estimated proportions may be used to determine the boundary position within the boundary region, as described above.

[0063] If the sum of the proportion of the boundary region occupied by the first object and the proportion of the boundary region occupied by the second object, when expressed as a percentage, is outside the predefined percentage range, this may indicate the presence of another object in the boundary region, the first reference region and / or the second reference region. For instance, in the example of Figure 2, the sheet of paper 270 cannot be distinguished from the desk 230 by the ToF sensor 210. This may lead to pixel 30 or 31 being identified as a first reference region. If the sheet of paper has sufficiently different reflective properties (for the wavelength used by the ToF sensor) to the desk, the ratio of the signal strength of the ToF signal for a boundary region for the first distance to the signal strength of the ToF signal for pixel 30 or 31 will not provide an accurate estimate of the proportion of the boundary region occupied by the first object.

[0064] Therefore, in response to a determination that the sum of the proportion of the boundary region occupied by the first object and the proportion of the boundary region occupied by the second object, when expressed as a percentage, is outside the predefined percentage range, the processing system 120 may be configured to identify a new first reference region and / or a new second reference region.

[0065] For instance, the processing system 120 may be configured to process the ToF signals 115 to identify a new first reference region and estimate a new proportion of the boundary region occupied by the first object based on a ratio of a signal strength of the ToF signal for the boundary region for the first distance to a signal strength of the ToF signal for the new first reference region. The new proportion may then be used as the proportion of the boundary region occupied by the first object when determining the boundary position within the boundary region.

[0066] The processing system 120 may be additionally or alternatively configured to process the ToF signals 115 to identify a new second reference region and estimate a new proportion of the boundary region occupied by the second object based on a ratio of a signal strength of the ToF signal for the boundary region for the second distance to a signal strength of the ToF signal for the new second reference region. The new proportion may then be used as the proportion of the boundary region occupied by the second object when determining the boundary position within the boundary region.

[0067] In some examples, the processing system 120 may be configured to identify a new first reference region and a new second reference region, and estimate new proportions of the boundary region for both the first object and the second object. The new proportions may then be used to determine the boundary position. In some examples, the sum of the new proportions may first be compared to the predefined percentage range, and used to determine the boundary position if the sum falls within the predefined percentage range.

[0068] In other examples, the processing system 120 may be configured to first identify a new first reference region and estimate a new proportion of the boundary region occupied by the first object. The sum of the new proportion of the boundary region occupied by the first object and the previously-estimated proportion of the boundary region occupied by the second object may be compared to the predefined percentage range. If the sum is still outside the predefined percentage range, a new second reference region may be identified and used to estimate a new proportion of the boundary region occupied by the second object.

[0069] In yet other examples, the processing system 120 may be configured to first identify a new second reference region and estimate a new proportion of the boundary region occupied by the second object. The sum of the new proportion of the boundary region occupied by the second object and the previously-estimated proportion of the boundary region occupied by the first object may be compared to the predefined percentage range. If the sum is still outside the predefined percentage range, a new first reference region may be identified and used to estimate a new proportion of the boundary region occupied by the first object.

[0070] In some examples, it may not be possible to identify first and second reference regions that result in a sum of proportions that falls within the predefined percentage range. This may be the case, for example, if the boundary region contains an object that is not covered (or not sufficiently covered) by any other region of the plurality of regions.

[0071] In such cases, the processing system 120 may be configured to determine the boundary position within the boundary region based on the boundary position within one or more other boundary regions containing a boundary between the first object and second object. For instance, if the sum of proportions falls within the predefined percentage range for other boundary regions on either side of the boundary region, the boundary position within the boundary region may be determined as a position in which the boundary connects the determined boundary positions within the other boundary regions.

[0072] Additionally or alternatively, in order to determine the boundary position within a boundary region for which proportions that sum to a value within the predefined percentage range cannot be achieved, the processing system 120 may be configured to repeat, after a predetermined interval, the estimation of the proportions of the boundary region occupied by the first object and second object respectively, using ToF signals acquired after the predetermined interval. This may allow the boundary position to be determined for a boundary region that temporarily contained an object other than the first and second objects (e.g. a sheet of paper on a desk may be removed after a few hours or days). The processing system may be configured to continuing repeating the estimation at predetermined intervals until proportions that sum to a value within the predefined percentage range are obtained.

[0073] In some examples, the processing system 120 may be configured to determine a boundary position for each of a plurality of identified boundary regions. The processing system may re-identify a first reference region and a second reference region for each boundary region. This may be the case, for example, where an adjacent or closest region containing only the first / second object is identified as the first / second reference region (e.g. in the example of Figure 2, the region corresponding to pixel 26 may be identified as the first reference region for the boundary region corresponding to pixel 34, while the region corresponding to pixel 27 may be identified as the first reference region for the boundary region corresponding to pixel 35). Alternatively, the same first reference region and second reference region may be used for more than one boundary region.

[0074] In some examples, where the processing system 120 is configured to determine the boundary position for each of a plurality of identified boundary regions, the determined boundary position for each boundary region may be based on the estimated proportions occupied by the first and second objects not only for that boundary region, but also for one or more other boundary regions. For example, the estimated proportions for one or more adjacent boundary regions may be used to determine the angle of the boundary position within a boundary region.

[0075] For instance, in the example of Figure 2, the estimated proportion occupied by the desk 270 is likely to be approximately the same across the boundary regions corresponding to the pixels 34 to 39. The boundary positions within these boundary regions may therefore be determined to be “straight” (i.e. have an angle of zero with respect to the row of pixels 34-39). If, on the other hand, a boundary region having an estimated proportion of 50% occupied by a first object was located between a boundary region having an estimated proportion of 30% occupied by the first object and a boundary region having an estimated proportion of 70% occupied by the first object, the boundary positions for these boundary regions may be determined to achieve a slanted boundary that passes through the boundary regions according to the estimated proportions for each region. In some examples, the identification of one or more other boundary regions may be used to determine a shape of the boundary region (e.g. to detect comers of boundaries). For instance, in the example of Figure 2, the identification of pixels 25 and 34 as corresponding to boundary regions indicates that the boundary within the boundary region for pixel 33 has a corner.

[0076] In some examples, having determined a boundary position within each of a plurality of boundary regions, the processing system 120 may be configured to process the determined boundary positions to generate a layout of the environment. For instance, in Figure 2, the dark grey shading in pixels 9-40 indicates the layout of the desk in the environment. In some examples (where the ToF signals indicate the presence of more than two objects in the environment), the processing system may be configured to determine boundary positions for additional objects in order to more fully determine the layout of the environment.

[0077] The layout of the environment may, for example, be used in the control of a smart lighting system or another smart appliance (e.g. a heating, ventilation and air conditioning, HVAC, system), in a security system or for healthcare monitoring. Further applications of a determination of a layout of an environment will be readily apparent to the skilled person.

[0078] Returning to Figure 1, in some examples, the system 100 further comprises a lighting device 150. The lighting device shown in Figure 1 is a desk lamp, but the skilled person will readily appreciate that any type of lighting device (e.g. floor lamp, ceiling lamp, etc.) may be used in the system 100. In some examples, the system 100 may comprise a plurality of lighting devices, each of which may be controlled by the processing system 120. For illustrative purposes, Figure 1 shows the ToF sensor 110 as separate to the lighting device 150; however, in some examples, the ToF sensor may be incorporated in a lighting device of the system 100.

[0079] The processing system 120 may use the determined boundary position(s) to control the lighting device 150. For instance, the processing system may be configured to process each determined boundary position to define a detection zone in the environment. The processing system may, continuously or at regular intervals, receive the ToF signals 115 for the plurality of regions from the ToF sensor 110, and process the ToF signals for the regions covering the defined detection zone to detect any movement in the detection zone. The processing system may be configured to control the lighting device 150 to provide light in response to detecting a movement in the detection zone. In the example shown in Figure 1, for instance, this would enable the lighting device 150 to be automatically switched on in response to a person approaching the desk 130.

[0080] The detection zone may be defined to end at a predetermined distance from the first or second object. For instance, in Figure 2, a detection zone extending 70 cm from desk 270 has been defined (illustrated by light grey shading in pixels 33-56). This detection zone only partially fills each of the regions corresponding to pixels 49-56. When a person is in one of these regions, the person’s position within the region may be estimated by detecting distance changes in the ToF signal for the region. In this way, a determination may be made as to whether the person is inside or outside the detection zone.

[0081] The examples provided above relate to a boundary between a desk and a floor; however, the skilled person will readily appreciate that the system 100 may be used in a variety of contexts. For instance, a boundary between a bed and a floor may be determined and used to control a night light (e.g. to provide light in response to detecting that a person is leaving the bed), or a boundary between a kitchen worktop and a floor may be used to control a lighting device provided above the worktop (e.g. to provide light in response to detecting that a person is approaching the worktop).

[0082] Figure 3 illustrates a computer-implemented method 300 for determining object boundaries, according to an embodiment of the invention.

[0083] The computer-implemented method 300 begins at step 310, at which a time- of-flight, ToF, signal is received, from a ToF sensor, for each of a plurality of regions of an environment. In some examples, each ToF signal may be acquired by a single pixel of the ToF sensor.

[0084] At step 320, the ToF signals are processed to identify a boundary region. The boundary region is a region of the environment containing a boundary between a first object, located at a first distance from the ToF sensor, and a second object, located at a second, different distance from the ToF sensor.

[0085] At step 330, the ToF signals are processed to identify a first reference region. The first reference region is a region of the environment that is predicted to contain only the first object. In some examples, the first reference region may be a region of the environment for which the ToF signal has a single peak corresponding to the first distance. In some examples, the first reference region may be adjacent to the boundary region.

[0086] At step 340, the ToF signals are processed to identify a second reference region. The second reference region is a region of the environment that is predicted to contain only the second object. In some examples, the second reference region may be a region of the environment for which the ToF signal has a single peak corresponding to the second distance. In some examples, the second reference region may be adjacent to the boundary region.

[0087] At step 350, a proportion of the boundary region occupied by the first object is estimated based on a ratio of a signal strength of the ToF signal for the boundary region for the first distance to a signal strength of the ToF signal for the first reference region.

[0088] At step 360, a proportion of the boundary region occupied by the second object is estimated based on a ratio of a signal strength of the ToF signal for the boundary region for the second distance to a signal strength of the ToF signal for the second reference region.

[0089] At step 380, the proportion of the boundary region occupied by the first object and the proportion of the boundary region occupied by the second object are processed to determine a boundary position for the boundary between the first object and the second object within the boundary region.

[0090] In some examples, the computer-implemented method 300 may further comprise a step 370, which may be carried out prior to step 380. At step 370, a determination is made as to whether a sum of the proportion of the boundary region occupied by the first object and the proportion of the boundary region occupied by the second object, when expressed as a percentage is within a predefined percentage range. In some examples, the predefined percentage range may have a lower limit between 80% and 90%, and an upper limit between 110% and 120%.

[0091] In response to a determination that the sum is within the predefined percentage range, the computer-implemented method may proceed to step 380.

[0092] In response to a determination that the sum is outside the predefined percentage range, steps 330 and 350 and / or steps 340 and 360 may be repeated. These steps may continue to be repeated until a sum of the most recently estimated proportion of the boundary region occupied by the first object and the most recently estimated proportion of the boundary region occupied by the second object is within the predefined range, at which point the computer-implemented method may proceed to step 380, determining the boundary position based on the most recently estimated proportion of the boundary region occupied by the first object and the most recently estimated proportion of the boundary region occupied by the second object.

[0093] In some examples, a plurality of boundary regions may be identified at step 320. The boundary position within each of the plurality of boundary regions may be determined by performing steps 330 to 380 for each boundary region. Alternatively, steps 330 and 340 may be performed once, and steps 350 to 380 may be performed for each boundary region, using the same first reference region and second reference region each time (except where step 370 is included and the sum falls outside the predefined range for a boundary region).

[0094] In some examples, the computer-implemented method 300 may further comprise a step (not shown in Figure 3) of processing each determined boundary position to generate a layout of the environment.

[0095] In some examples, the computer-implemented method 300 may further comprise the steps (not shown in Figure 3) of: processing each determined boundary position to define a detection zone in the environment; processing the ToF signal for each region of the environment in the detection zone to detect, if present, a movement in the detection zone; and controlling a lighting device to provide light in response to detecting a movement in the detection zone.

[0096] It will be understood that the disclosed methods are computer-implemented methods. As such, there is also proposed a concept of a computer program comprising code means for implementing any described method when said program is run on a processing system.

[0097] As discussed above, embodiments make use of a controller. The controller can be implemented in numerous ways, with software and / or hardware, to perform the various functions required. A processor is one example of a controller which employs one or more microprocessors that may be programmed using software (e.g., microcode) to perform the required functions. A controller may however be implemented with or without employing a processor, and also may be implemented as a combination of dedicated hardware to perform some functions and a processor (e.g., one or more programmed microprocessors and associated circuitry) to perform other functions.

[0098] Examples of controller components that may be employed in various embodiments of the present disclosure include, but are not limited to, conventional microprocessors, application specific integrated circuits (ASICs), and field-programmable gate arrays (FPGAs).

[0099] In various implementations, a processor or controller may be associated with one or more storage media such as volatile and non-volatile computer memory such as RAM, PROM, EPROM, and EEPROM. The storage media may be encoded with one or more programs that, when executed on one or more processors and / or controllers, perform the required functions. Various storage media may be fixed within a processor or controller or may be transportable, such that the one or more programs stored thereon can be loaded into a processor or controller.

[0100] Variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure and the appended claims. In the claims, the word "comprising" does not exclude other elements or steps, and the indefinite article "a" or "an" does not exclude a plurality.

[0101] Functions implemented by a processor may be implemented by a single processor or by multiple separate processing units which may together be considered to constitute a "processor". Such processing units may in some cases be remote from each other and communicate with each other in a wired or wireless manner.

[0102] The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage.

[0103] A computer program may be stored / distributed on a suitable medium, such as an optical storage medium or a solid-state medium supplied together with or as part of other hardware, but may also be distributed in other forms, such as via the Internet or other wired or wireless telecommunication systems.

[0104] If the term "adapted to" is used in the claims or description, it is noted the term "adapted to" is intended to be equivalent to the term "configured to". If the term "arrangement" is used in the claims or description, it is noted the term "arrangement" is intended to be equivalent to the term "system", and vice versa.

[0105] Any reference signs in the claims should not be construed as limiting the scope.

Claims

CLAIMS:

1. A computer-implemented method (300) for determining object boundaries, the computer-implemented method comprising: receiving, from a time-of-flight, ToF, sensor (110, 210), a ToF signal (115) for each of a plurality of regions of an environment; processing the ToF signals to identify a boundary region, wherein the boundary region is a region of the environment containing a boundary between a first object (130, 230), located at a first distance from the ToF sensor, and a second object (140, 240), located at a second, different distance from the ToF sensor; processing the ToF signals to identify a first reference region, wherein the first reference region is a region of the environment that is predicted to contain only the first object; processing the ToF signals to identify a second reference region, wherein the second reference region is a region of the environment that is predicted to contain only the second object; estimating a proportion of the boundary region occupied by the first object based on a ratio of a signal strength of the ToF signal for the boundary region for the first distance to a signal strength of the ToF signal for the first reference region; estimating a proportion of the boundary region occupied by the second object based on a ratio of a signal strength of the ToF signal for the boundary region for the second distance to a signal strength of the ToF signal for the second reference region; and processing the proportion of the boundary region occupied by the first object and the proportion of the boundary region occupied by the second object to determine a boundary position for the boundary between the first object and the second object within the boundary region.

2. The computer-implemented method (300) of claim 1, wherein each ToF signal (115) is acquired by a single pixel of the ToF sensor (110, 210).

3. The computer-implemented method (300) of claim 1 or 2, wherein: the first reference region is a region of the environment for which the ToF signal (115) has a single peak corresponding to the first distance; and the second reference region is a region of the environment for which the ToF signal has a single peak corresponding to the second distance.

4. The computer-implemented method (300) of any of claims 1 to 3, wherein the first reference region and / or the second reference region are adjacent to the boundary region.

5. The computer-implemented method (300) of any of claims 1 to 4, further comprising, in response to a determination that a sum of the proportion of the boundary region occupied by the first object (130, 230) and the proportion of the boundary region occupied by the second object (140, 240), when expressed as a percentage, is outside a predefined percentage range: processing the ToF signals (115) to identify a new first reference region; estimating a new proportion of the boundary region occupied by the first object based on a ratio of a signal strength of the ToF signal for the boundary region for the first distance to a signal strength of the ToF signal for the new first reference region; and processing the new proportion of the boundary region occupied by the first object to determine the boundary position within the boundary region.

6. The computer-implemented method (300) of any of claims 1 to 5, further comprising, in response to a determination that a sum of the proportion of the boundary region occupied by the first object (130, 230) and the proportion of the boundary region occupied by the second object (140, 240), when expressed as a percentage, is outside the predefined percentage range: processing the ToF signals (115) to identify a new second reference region; estimating a new proportion of the boundary region occupied by the second object based on a ratio of a signal strength of the ToF signal for the boundary region for the second distance to a signal strength of the ToF signal for the new second reference region; and processing the new proportion of the boundary region occupied by the second object to determine the boundary position within the boundary region.

7. The computer-implemented method (300) of claim 5 or 6, wherein the predefined percentage range has a lower limit between 80% and 90%, and an upper limit between 110% and 120%.

8. The computer-implemented method (300) of any of claims 1 to 7, further comprising: processing the ToF signals (115) to identify one or more further boundary regions, wherein each further boundary region is a region of the environment containing a boundary between the first object (130, 230) and the second object (140, 240); and for each further boundary region: estimating a proportion of the further boundary region occupied by the first object based on a ratio of a signal strength of the ToF signal for the further boundary region for the first distance to a signal strength of the ToF signal for the first reference region; estimating a proportion of the further boundary region occupied by the second object based on a ratio of a signal strength of the ToF signal for the further boundary region for the second distance to a signal strength of the ToF signal for the second reference region; processing the proportion of the further boundary region occupied by the first object and the proportion of the further boundary region occupied by the second object to determine a boundary position for the boundary between the first object and the second object within the further boundary region.

9. The computer-implemented method (300) of claim 8, wherein the determination of each boundary position is further based on the proportion occupied by the first object (130, 230) and the proportion occupied by the second object (140, 240) for one or more other boundary regions.

10. The computer-implemented method (300) of claim 8 or 9, further comprising processing each determined boundary position to generate a layout of the environment.

11. The computer-implemented method (300) of any of claims 1 to 10, further comprising: processing each determined boundary position to define a detection zone in the environment;processing the ToF signal (115) for each region of the environment in the detection zone to detect, if present, a movement in the detection zone; and controlling a lighting device (150) to provide light in response to detecting a movement in the detection zone.

12. A computer program product comprising computer code means which, when executed on a computing device having a processing system, cause the processing system to perform all of the steps of the method (300) according to any of claims 1 to 11.

13. A processing system (120) for determining object boundaries, the processing system being configured to: receive, from a time-of-flight, ToF, sensor (110, 210), a ToF signal (115) for each of a plurality of regions of an environment; process the ToF signals to identify a boundary region, wherein the boundary region is a region of the environment containing a boundary between a first object (130, 230), located at a first distance from the ToF sensor, and a second object (140, 240), located at a second, different distance from the ToF sensor; process the ToF signals to identify a first reference region, wherein the first reference region is a region of the environment that is predicted to contain only the first object; process the ToF signals to identify a second reference region, wherein the second reference region is a region of the environment that is predicted to contain only the second object; estimate a proportion of the boundary region occupied by the first object based on a ratio of a signal strength of the ToF signal for the boundary region for the first distance to a signal strength of the ToF signal for the first reference region; estimate a proportion of the boundary region occupied by the second object based on a ratio of a signal strength of the ToF signal for the boundary region for the second distance to a signal strength of the ToF signal for the second reference region; and process the proportion of the boundary region occupied by the first object and the proportion of the boundary region occupied by the second object to determine a boundary position for the boundary between the first object and the second object within the boundary region.

14. A system (100) for determining object boundaries, the system comprising: a time-of-flight, ToF, sensor (110, 210); and the processing system (120) of claim 13.

15. The system (100) of claim 14, further comprising a lighting device (150), wherein the processing system (120) is configured to control the lighting device based on the determined boundary position.

Images