ES2540236T3 - Apparatus and method to inspect and classify a product flow - Google Patents

Abstract

Apparatus for classifying a product flow (2, 3), comprising: - a supply system (1) for supplying the product flow to a scanning area (28); - a background element (5) located behind the scanning area; - at least one inspection unit (9, 10) comprising: - a scanning unit (43) to illuminate the product flow along the entire width W of the product flow, and - a detection unit (44 ) to capture an area of light returned by the products; - a rejection system (11); characterized in that said background element (5) consists of a means for capturing incident light and for directing it to detection means (40, 40 ') that are configured to convert said light into an electrical signal, said apparatus further comprising a means to generate a location signal (C ') that is indicative of the location of the products, by detecting the immediate incidence of light on the background element (5), in which said detection unit (44) is a camera arranged to capture an image of the scanning area (28).

Description

DESCRIPTION

Apparatus and method to inspect and classify a product flow

Field of the Invention

The present invention relates to a method and an apparatus for inspecting a product flow. This inspection can also lead to a quality selection through a classification stage. 5

The invention is applicable when certain degraded products and foreign objects are removed from an incoming product flow.

The invention is particularly suitable for the classification of food products such as green beans, peas, nuts, raisins, cauliflowers, lettuces and the like, among which non-food products, such as wood, plastic, glass and others, must be removed from the flow of products. 10

The invention is also extraordinarily suitable for the classification of non-food products, such as plastics in garbage recycling, glass classification and the like.

Prior art

It is known from the international patent application WO 01/00333 that product objects in a product flow can be illuminated with a concentrated beam of light. The reemitted light is captured by a detector after 15 of which it is analyzed. Based on this analysis, a selection mechanism can be controlled to achieve a particular classification result.

In the absence of a product, the light is reflected by a background element that needs to be chosen depending on the product. More specifically, the fund must be chosen in such a way that it shows the optical characteristics of a good product. That is, the good product is invisible against the background. However, the deviations in an analyzed product, such as insect bites, putrescent spots, foreign objects and the like, produce a deviation in the returned light signal. By setting a threshold, the difference can be made between the light signal coming from the bottom and acceptable products, on the one hand, and the light signal coming from an unacceptable product that must be removed, on the other hand.

Similar background elements are described in US Patent 4,723,659 and in European Patents 25 EP1012582 and EP0443469. In general terms, they refer to a background element located perpendicular to the direction of product movement, in the field of view of the detection means. Generally, the bottom element is a cylindrical roller. While turning and with the help of a scraper it becomes self-cleaning.

A disadvantage of this method is that a specific background element must be available for each type of product. In this way, a background element for carrots should have an orange color, while 30 for green beans should have a green color. During the classification of food products and, more specifically, while changing from one product to another, it is also necessary to change the background element. In addition, the cost of a fund element is not negligible.

In certain applications, other optical properties are provided to the background element. Many fresh vegetables contain, for example, chlorophyll. A frequency shift to the infrared spectrum occurs within these molecules when illuminated with light that has a wavelength between 640 and 680 nanometers. This emission phenomenon is called fluorescence. By giving the background element the same fluorescent properties, in the prior art, it can also be classified based on the presence of chlorophyll in the analyzed products. Other molecules, such as aflatoxin, also show fluorescent properties, and in principle, could be detected in the products analyzed according to a similar technique. 40

An important disadvantage is that, over time, these types of background elements lose their fluorescent properties. This has adverse effects on the quality of classification and is not interesting for users of this type of equipment.

For these reasons, there is a need for a classification device, in particular an inspection device, without the disadvantages of the devices known in the prior art. Four. Five

For these reasons, there is a need for a method that does not show the disadvantages of the current state of the art.

 Objects of the invention

In general, the object of the invention is a method and an apparatus for carrying out a selection between products in a large continuous product flow in a very efficient, reliable and cost-effective manner.

In each of WO-A-2008/142496 or EP 1724030, an apparatus according to the preamble of claim 1 and a method according to the preamble of claim 9 are described.

More specifically, according to a number of preferred embodiments, the object of the invention is an inspection method and inspection means that prevent the replacement of the background element during a product change.

That is, the object of the invention is an inspection method and inspection means that use a background reference that is at least independent of the type of product.

Summary of the Invention

The invention relates to an apparatus for classifying a product flow, comprising:

 a supply system 1 for supplying the flow of products to a scanning area,

 a background element located behind the scanning area, 10

 at least one inspection unit consisting of

or a scanning unit to illuminate the product flow along the entire width W of the product flow, and

or a detection unit to capture an area of light returned by the products,

 a rejection system, 15

wherein said background element consists of means for capturing incident light and directing it to a detection means that is configured to convert said light into an electrical signal, said apparatus further comprises means for generating a location signal that is indicative of the location of the products, by detecting the immediate incidence of light on the background element, in which said detection unit is a camera arranged to capture an image of the scanning area. twenty

According to one embodiment, the background element is an optical fiber with a grooved surface to capture incoming light.

According to one embodiment, the background element is a plurality of detectors configured to convert the incident light into an electrical signal.

The invention further relates to a method for classifying a flow of products in products to be accepted and in 25 products to be rejected, comprising the steps of:

• move the products to be classified through a scanning area, supplied in a product flow that extends over a given width and has a thickness of substantially a single layer of products,

• in this scanning area, scan linearly across the width of this product flow using one or more concentrated light beams, which illuminate, in the absence of products, a background element placed behind this product flow and that extends over the width of it,

• capture a camera image of said scanning area, and deduce from said image, electrical signals representative of each product in said image,

• generate one or more control signals based on said image, whereby these control signals allow a selection of among the scanned products to be accepted, on the one hand, and the scanned products to be rejected, on the other hand, and

• classify the flow of products by means of these one or more control signals, characterized in that

• the background element consists of a means to capture the incident light and direct it towards a detection means that is configured to convert said light into an electrical signal, and

• said step of generating one or more control signals comprises:

• generate a signal that is indicative of the location of the scanned products, in which the generation of the location signal comprises detecting and converting that part of the concentrated scanning light beam, which is not obstructed by the products, thus obtaining a signal that is indicative of the location of the scanned products,

• generate said control signals depending on whether the electrical signals representative of said products and that derive from said camera image cross or not a threshold in the areas where a product is present according to said signal indicative of the location of the scanned products.

According to one embodiment, the background element is an optical fiber with a grooved surface to capture incoming light.

According to one embodiment, the background element is a plurality of detectors configured to convert the incident light into an electrical signal.

The method of classifying products according to the invention is characterized at least by transporting the 5 products to be classified along a given path in the form of a product flow having a width; by a bottom element that extends along the width of the product flow; with a concentrated beam of light illuminating, along the width of the product flow, the products to be classified and, in the absence of said products, the background element; for capturing the light reissued by the products and the background element; by carrying out, based on said observed light, a first selection between the background element, on the one hand, and all the products in the product flow, on the other hand; for carrying out a second selection of the products to be classified, on the one hand, and the products to be rejected, on the other hand, and based on this second selection, to automatically carry out a separation of the products in said product flow.

Preferably, the bottom element comprises a surface that extends along the width of the product flow, whereby said surface reflects at least partially incoming light. fifteen

In a particularly useful embodiment, the bottom element is in the form of a cylindrical roller.

In an alternative embodiment, the background element comprises means for capturing and redirecting incoming light towards an optical-electric converter. In this case, such a background element generates a signal that has a progression from which the presence or absence of products in the scanning zone can be easily deduced. A Boolean signal thus obtained is particularly useful in the method according to the invention. twenty

The first selection is preferably made depending on whether the intensity of the detected light or a signal derived from it crosses a threshold value or not.

In some cases, more specifically when the total product range includes positive and negative peaks against the background signal, the first selection is made based on whether the intensity of the observed light emitted by the background element or a signal derived from the it falls within a zone, or not, said zone being further characterized by a maximum threshold value.

The second selection is preferably made depending on whether the intensity of the detected light or a signal derived from it crosses a threshold value, or not.

In a preferred embodiment, said crossing of a threshold value is defined exclusively within those zones that are marked during the first selection stage as coming from the product. 30

In an exceptionally preferred embodiment, a new signal is generated after the first selection, which is further characterized by the preservation of the intensity of the light observed in those areas that delimit the product (the product zones) and, subsequently, by the change of intensity in areas where the background element can be observed at another level.

Furthermore, it is preferable to filter said signal so that the high frequency transitions in the product zones 35 are flattened and a new signal is created. Said filter could be, for example, an adaptive filter tuned specifically to smooth the transition from the product zone to the bottom zone and vice versa.

In the most practical embodiment according to the invention, the second selection is made in said filtered signal.

In any case, the selection of the background element will be such that it gives rise to at least one corresponding signal having a path according to which a first selection can be made between said background element, 40 on the one hand, and all products in the product flow, on the other hand.

In a practical embodiment of the method according to the invention, scanning is performed using a rotating mirror, preferably a rapidly rotating polygonal mirror.

In a very practical embodiment, the scan uses a laser beam.

In a practical embodiment, the products are transported on a vibrating table, tape or the like, to an inspection facility.

In some cases, more specifically in the case of free fall sorting devices, the products are also guided during their free fall by a free fall plate. In addition, the products to be separated are separated.

by a manifold of air valves located along the width of the product flow and which are opened based on the second selection stage.

In some cases, more specifically in those cases where the defects are located on both sides, it is advantageous to scan the products to be classified from two sides, one opposite the other.

The method could be combined with color classification when performing classification based on light reflections. 5

In addition, different concentrated light beams could be used, each having a different wavelength, possibly combined into a single beam.

In an important variant according to the invention, two signals are combined in a two-dimensional graph, such that each point in this graph corresponds to a specific intensity level according to the path of the first signal combined with a specific intensity level according to the trajectory of the second signal; the points corresponding to the product to be accepted are grouped in the first zone; the points that correspond to the products to be rejected are grouped into a second zone; the points corresponding to the background element are delimited by a third zone; The background signal level is adjusted by repositioning the third zone to a new location.

In this case, moving said third zone can be achieved by displaying this zone in a graph 15 shown in a graphical user interface and subsequently dragging this zone to a new location.

In a preferred embodiment, said new location is chosen so that a separation can be made between the first and the third zone, on the one hand, and the second zone, on the other hand, using a separation plane.

In addition, more than two signals can be combined in a larger dimension chart.

Apart from said method, the invention also relates to an apparatus for classifying products using this method and in such a way that it comprises at least one transport device for transporting a flow of products, which extend over a width, in a specific direction; means for scanning the products to be classified along the width of the product flow, which further comprises means for generating a concentrated light and directing it towards the means for emitting this beam of light on the products; means to capture the return light; means for carrying out a selection of the products scanned based on the observed light; 25 means for separating the products based on this selection.

In a preferred embodiment, the means for generating the concentrated light is a laser generator.

In a preferred embodiment, the means for emitting the light beam on the products comprise optical means, more specifically a rotating polygonal mirror, which moves the concentrated light transversely through the product flow. However, the present invention is not limited to such a scanning arrangement. One could generate, by way of example, a row of concentrated beams of light, which are possibly connected and disconnected in sequence.

In addition, the means for carrying out a selection based on the return light could be based on digital electronic components, more specifically on Field Programmable Door Arrays and microprocessors, or could be based on analog electronic components such as amplifier circuits. operational, or could be a combination of analog and digital processing units.

In a practical embodiment, the means for making a separation between the products based on said selection are formed by an air valve manifold, mounted transversely through the product flow.

In an advantageous embodiment, the bottom element is formed by a surface across the width of the product flow, such that the incident light is at least partially reflected by said surface. 40

In an alternative embodiment, the background element is formed by means that capture and channel the incident light to the means to convert this light into an electrical signal.

In a preferred embodiment, the means for capturing the return light are formed by an optical filter that makes the detection arrangement sensitive for a specific spectrum of light, in operative communication with a spatial filter that makes the detection arrangement sensitive to a specific area of the return light, and in operational communication with both filters, an optical-electric converter transforms the light into a corresponding electrical signal.

A method for classifying a product flow into products to be accepted and products to be rejected comprises the steps of moving the products to be classified through a scanning area, which are supplied in a product flow that covers a certain width and which has a thickness of substantially a single layer of 50

products, scan linearly in this scanning area one or more beams of light concentrated across the width of this product flow, illuminate, in the absence of products, a background element located behind this product flow that extends over its width, whereby this beam of light produces light signals in these scanned products and in this scanned background element, detect these light signals whereby these light signals are converted into electrical signals, generate one or more control signals based on these converted signals whereby these control signals allow a selection of among the scanned products to be accepted, on the one hand, and the scanned products to be rejected, on the other hand, and to classify the flow of products by means of these one or more control signals, characterized in that this method further comprises; choose this background element so that the corresponding detected light signal differs by at least 1 parameter of the light signals of the products to be classified and because the generation of one or more control signals comprises further changing the background level. of the converted signals after the detection of the light signals, towards a new level of the chosen signal such that, in the signal obtained in this way, the signal level of the signal of a scanned product to be accepted differs from Signal level of a scanned product to reject. The parameter in which the background element can be distinguished from the products to be classified may be the level of the signal, a spatial aspect such as dispersion or a frequency aspect such as color information.

This method may further comprise, after moving the background level of the converted signals, comparing the signal obtained in this way with one or more threshold values to thereby generate the one or more control signals.

Moving the background level according to this method may further comprise generating a signal that is indicative of the location of the scanned products in the detected and converted signals and changing the level of the converted signals to locations other than those of the products. scanned, as indicated by this location signal.

Generating the location signal according to this method may further comprise the detection and conversion of the light signals from the scanned background element and the scanned products, into these 25 converted signals that separate the signal from the element of Scanned background of the signals of the scanned products in such a way that a signal indicative of the location of the scanned products is obtained.

Distinguishing the signal from the background element scanned according to this method may further comprise comparing the converted signals with one or more threshold values. 30

An alternative embodiment of these methods of generating the location signal may further comprise the detection and conversion of that area of the concentrated scan beam passing through the products, thereby achieving a signal that is indicative of the location of scanned products.

The location signal in the different embodiments of this method may be a Boolean signal. The signal level shift to a new signal level can then occur according to the formula D = 35 BC + s (C⊕1), in which B are the detected and converted light signals, s is a real number chosen based on the desired displacement of the signal level, C is the location signal, and ⊕ is defined as the addition of module - 2.

The location signal in the different embodiments of this method may be a Boolean signal. The signal level shift to a new signal level can then occur according to the operation D = B when C = 1 and D = 0 when C = 0 and where B is the detected light signal, and C is The location signal. 40

The location signal in the different embodiments of this method may be an analog or digital signal, so that the comparison of the signal obtained in this way with one or more threshold values occurs only at the location of the scanned products as indicated by The location signal.

The background level of the converted signals can be shifted to a signal level according to that of a product to be accepted. Four. Five

The location signal in any embodiment can be generated based on the one or more first signals detected and then can be used to indicate the location of the products in one or more second signals detected.

The background element in any of the embodiments may consist of a surface that extends over the width of the product flow, whereby said surface reflects, at least partially, the incident light. Preferably, this bottom element is in the form of a cylindrical roller.

In an alternative embodiment, this background element consists of means for capturing and redirecting the incident light towards an optical-electric converter. In this case, the background element may be an optical fiber with a grooved surface to capture the incoming light. Such optical fiber may consist of segments, so for each

segment the orientation of these slots is chosen with respect to the longitudinal direction of the optical fiber as a function of the position of this segment along the width of the product flow.

In any embodiment according to this method, the signal thus obtained can be filtered before, so that the high frequency transitions near the product areas are flattened, generating a new filtered signal. This filter can be an adaptive filter that specifically adjusts to smooth transitions from a product zone to a background zone and vice versa.

In any embodiment according to this method, linear scanning of the flow of products can be performed by means of a moving mirror, preferably a fast-rotating polygonal mirror. The product flow can be scanned by several concentrated light beams by means of this moving mirror, so that each light beam has a separate frequency. Preferably, this concentrated beam of light is a laser beam. The products to be classified can be scanned from both edges of the product flow.

In any embodiment of this method, the flow of products can be supplied by means of a vibrating table, a conveyor belt or the like. The products can also be supplied by means of a free fall plate that guides these products during their free fall towards the scanning area.

In any embodiment of this method, the classification of the flow of products by means of these one or more 15 control signals can be produced by controlling an air valve manifold located across the width of the flow of products by means of of these one or more control signals.

In any embodiment, in addition to doing it at the signal level, the classification of the product flow can also occur by color, that is, the frequency of the detected light signals.

In an embodiment of the aforementioned methods in which the generation of one or more control signals 20 by moving the background level of the converted signals to a new signal level that is chosen as such so that, in the signal obtained in this way, the signal level of the signal of a scanned product to be accepted differs from the signal level of the signal of a scanned product to be rejected, which further comprises; combine two signals detected and converted into a two-dimensional graph, in which each point corresponds to a particular intensity level according to the trajectory of the first signal combined with a particular intensity level according to the trajectory of the second signal; the points that correspond to the product to be accepted are grouped in a first zone; the points that correspond to the product to be rejected are grouped into a second zone; the points that correspond to the background element are delimited by a third zone; The background signal level is adjusted by repositioning the third zone to a new location. Moving said third zone can be done by visualizing this area in a graphic that is displayed in a graphical user interface and subsequently dragging this area to a new location. The mentioned new location can be chosen in such a way that a separation can be made between the first and the third zone, on the one hand, and the second zone, on the other hand. This two-dimensional graph may have an additional dimension that shows the histogram of combinations of signals that appear. In addition, more than two detected signals can be combined with each other in graphics with more dimensions so that, for each location in the scanning area, as much optical information as possible can be collected and represented, allowing a better distinction to be made. between the location of the products to be classified and those of the fund, on the one hand, and between the products to be accepted and the products to be rejected, on the other. All or at least one of said first, second and third zones in such two or more dimensional graphs can be deduced by automatic grouping algorithms. 40

An apparatus for classifying products according to the methods of any of the preceding claims, is characterized in that at least it consists of a delivery system that transports a single layer of the products to be classified along a determined path in the form of a product flow that extends over a width; means for scanning the products to be classified by the width of this product flow, whereby these scanning means further comprise; means for generating a concentrated beam of light and directing it towards the 45 products through optical means; means for detecting the return light and converting it into an electrical signal; means for generating control signals that allow the selection of among the scanned products based on said detected light; and means for classifying the flow of products based on said selection by means of said one or more control signals, characterized in that the classification apparatus further comprises: a background element chosen in such a way that the light signal correspondingly detected differs by at least 1 50 parameter of the light signals of the products to be classified and in that the selection means comprise means for generating one or more control signals by shifting the background level of the light signals towards a signal level chosen in such a way that, in the signal obtained in this way, the signal level of a scanned product to be accepted differs from the signal level of a scanned product to be rejected.

The selection means of such a classification device may further comprise means for generating a location signal based on the one or more detected signals, means for obtaining a signal based on this location signal and based on these or other or more signals detected, such that the background level in these

Last signals are moved to a new level that allows differentiating the products to be accepted from the products to be rejected in said signal obtained.

These selection means may further comprise means for comparing the signal obtained with one or more threshold values, thereby generating the one or more control signals.

These selection means may further comprise filters to prefilter the signal thus obtained, so that the 5 high frequency transitions near the product areas are flattened and thus obtain a new filtered signal. This filtering operation can be achieved by means of an adaptive filter specifically adjusted to smooth transitions from a product zone to a bottom zone and vice versa.

In any of the aforementioned sorting apparatus, this background element may consist of a surface that extends over the width of the product flow, whereby said surface reflects the light at least partially. Preferably, this bottom element is in the form of a cylindrical roller.

In an alternative embodiment, this background element may consist of means for capturing and redirecting the incident light towards an optical-electric converter. Such a background element may be an optical fiber with a grooved surface to capture the incident light. Such optical fiber may consist of segments, so that in each segment the orientation of these grooves is chosen with respect to the longitudinal direction of the optical fiber as a function of the position of this segment along the width of the product flow. .

A laser can be used in such sorting devices to capture the concentrated beam of light. This laser can be moved across the width of the product flow by means of a rotating polygonal mirror.

In such sorting devices, the means for making a selection from among the products scanned according to the observed light, consist of a signal processing platform based on digital electronic components 20, more specifically Field Programmable Door Arrays and microcomputer processors, or based on analog electronic circuits, such as operational amplifier circuits, or a combination of analog and digital processing units.

In such sorting apparatus, the means for classifying the flow of products according to said selection by means of these one or more control signals, consist of an air valve manifold, positioned transversely through the flow of products.

In such sorting devices, the means for capturing the light may consist of an optical filter that makes the detector sensitive to a particular spectrum of light; in operative communication with a spatial filter that makes the detector sensitive to a particular area of the return light; in operational communication with both of the aforementioned filters and optical-electric converter that converts the light into a corresponding electrical signal. 30

Brief description of the figures

Figure 1 schematically illustrates the main operation of a sorting apparatus according to the invention;

Figure 2 schematically illustrates a possible embodiment of a scanning device;

Figures 3a and 3b show alternative embodiments of the background element; 35

Figure 4 schematically illustrates a detection unit;

Figure 5 schematically illustrates an apparatus with several detection units;

Figures 6 a-e illustrate the multi-stage method according to the present invention that results in better inspection or at least more advantageous;

Figures 7 a-b illustrate this method in a two-dimensional representation; 40

Figure 8 shows a classification apparatus according to the invention in the viewing perspective;

Figures 9 a-f schematically illustrate the signal processing in an inspection device when the background element generates a signal that deviates from that of a good product;

Figures 10 a-d schematically illustrate the signal processing according to an embodiment of the invention; Four. Five

Figures 11 a-d schematically illustrate the signal processing according to an embodiment of the invention;

Figures 12 a-d schematically illustrate the signal processing according to an embodiment of the invention;

Figures 13 a-c schematically illustrate the signal processing according to an embodiment of the invention;

Figure 14 schematically illustrates the signal processing according to an embodiment of the invention;

Figures 15 a-c schematically illustrate the signal processing according to an embodiment of the invention.

Detailed Description of the Invention 10

The present invention will be described by means of some examples, which refer to certain figures, without any restriction. The figures are only schematic and not limiting. In the figures, the dimensions of certain elements may have been exaggerated or may not be in real proportion. This is due to illustrative reasons. For this reason, the relative dimensions and dimensions do not necessarily correspond to reality.

The present invention describes a method and mechanism for classifying products 2, 3, to be precise, granular products such as raisins, beans, berries, but also plastic grains, which are transported in large quantities and in a continuous flow.

In addition, the method according to the invention is also suitable for the inspection of larger products, such as beans, cauliflowers, lettuces, etc.

Figure 1 schematically shows a sorting apparatus 14 according to the present invention. This sorting apparatus contains a supply system 1, at least one inspection unit 9, 10, a rejection system 11 and possibly a free fall ramp 4 that guides the flow of products in free fall to the storage unit. inspection 9, 10 and the rejection system 11. The ramp, however, can also be a conveyor belt that transports products 2, 3.

The delivery system 1 with a width W can be a vibrating table, or any other transport system 25 known in the current state of the art. In case the supply system 1 is functioning as a conveyor belt, the use of a free fall ramp 4 can be superfluous, as is well known to one skilled in the art.

In the scanning area 28, the inspection unit observes the fall of the product 2, 3, by analyzing the return light. Based on this analysis, a rejection system 11 is controlled. This results in a separation of the product flow into an acceptance flow 13 and a rejection flow 12.

Behind the scanning area 28, a background element 5 is illuminated and observed while no product 2, 3 is present in that area 28. The optical properties of the background element 5 are chosen in such a way that a proper distinction between all products 2, 3 in the product flow, on the one hand, and the background element, on the other hand. These optical properties may refer to the frequency or the spatial properties of the background element 5. The background element may generate a light signal with another frequency or reflect the incident light 34 in another way, or even disperse it. This method differs from the current state of the art in that it tries to make a distinction between all the products to be rejected 3 of the product flow, on the one hand, and the fund 5 together with the products to be accepted 2, on the other hand. For this reason, in the present invention the choice of the bottom element 5 becomes considerably less complicated and independent of the type of product. For example, background element 5 will be the same for green beans and orange carrots. This differs from the current state of the art, in which a specific background element has to be available for each product that has identical optical properties to those of products 2 to be accepted. In the case of green beans and carrots, in the current state of the art two different background elements are needed: one with a green color, and another with an orange color. When the products to be classified 2, 3 are transported by a conveyor belt 1, this conveyor can be used as a background element to obtain a background signal that differs from the signal of all products 2, 3, as described in the embodiments of the invention.

In a preferred embodiment of the present invention, the bottom element 5 is a cylindrical roller, to be precise, a rotating roller that cleans itself by means of a scraper that is placed against the roller. fifty

In a preferred configuration for the classification of fresh vegetables, such as green beans, carrots and peas, the bottom element 5 is applied as a white cylindrical roller, highly diffuser, not fluorescent. With

reference to figures 4, 5 and 6: a signal 39, measured by a detection unit 44, which is only sensitive to scattered light, in this case it will show a path B along which a suitable distinction can be made between the bottom signal 19, on the one hand, and the peaks 20, 21 from green beans, peas, carrots, wood, plastic, metal, glass, etc., on the other.

In an alternative version, the bottom element 5 is an intrinsic component of the mechanical construction and has, in addition to the optical background function, an additional function as a mechanical support element. As a result, it cannot be withdrawn as such. Such a simplified mechanical embodiment to a large extent according to the present invention is impossible to apply in the current state of the art.

The inspection unit 9, 10 consists of a scanning unit 43 and a detection unit 44 (Figures 4 and 5). The scanning unit illuminates the flow of products along the entire width W. The detection unit 44 10 captures an area of the return light and will convert this return light by at least one optical-electrical converter 38, into a electrical signal that is then analyzed in a processing unit 41.

Figure 2 illustrates a possible embodiment of a scanning unit 43. A concentrated beam of light 45, preferably from a laser 29, is directed towards a fast rotating reflective polygonal wheel 30. During rotation, this polygon 30 generates a point of concentrated light that moves rapidly directed towards the flow of products. The scanning is carried out along the entire width W with a scanning angle determined by extreme light beams 31, 31 '.

When this light 32, 33, 34 falls on a product 2, 3 an area of the light 46, 47, 48 will be reflected according to the color of this product 2, 3. This makes possible the color classification.

Depending on the light permeability of the illuminated product 2, 3, the concentrated light beam 32, 33, 34 will be directly reflected and / or dispersed, as described in detail in US Patent 4,634,881. This makes possible the classification by structure. In addition, the presence of fluorescent molecules in the product 2, 3 will cause a frequency shift in the reflected light, so it is possible to classify by the presence of those molecules, such as chlorophyll and aflatoxin.

Completely within the scope of this patent, several light beams with different wavelengths can be grouped, preferably by combining several lasers of different wavelengths by means of mirrors and optical filters.

In Figures 3a and 3b, an alternative version of the background element 5 is clarified, in which this element consists of means to capture the incident light 34 and to direct it to a detector 40 that converts this light into an electrical signal 39. 30

As illustrated in Figure 3a, the bottom element 5 may consist of an optical fiber 56, in which, by means of a specific mechanism closer to it 57, the incident light 34 can also be channeled to a detector 40 that generates a electrical signal 39. The element 57, for example, may consist of slots or small mirrors connected to the fiber 56, so that the light bends towards the aforementioned filter 56.

In the international patent application WO 2007/062219 and in part 2 entitled "Design and manufacturing", of article 35 "A light source distributed based on a fiber network", by GE Carver, Proc. of SPIE volume 6371, 63710H-2 (2006), both of which are integrally included in this description, a slotted optical fiber is used to obtain a uniform linear light source. The light injected into the fiber optic in its longitudinal direction is redirected in a direction that differs from the longitudinal one through grooves on the fiber side. Redirected light can also be guided through a cylindrical optical system to obtain a more uniform light distribution within a restricted area. According to the choice of the geometric parameters of the grooves (see Figure 5 of WO 2007/062219 or Figure 2 of the SPIE article: width, step d, angle of the groove in relation to the normal perpendicular direction in the direction longitudinal, the angle of the groove in relation to the longitudinal direction) and the optical parameters (wavelength of the light source, refractive index of the optical fiber), the direction and the degree of redirection can be determined. The slots can be applied by means of a printer using laser light or UV light (see chapter 2, second paragraph of the SPIE article, page 9, third paragraph, which is illustrated in Figure 5 of WO 2007/062219) . By varying the relative orientation of the optical fiber with respect to the printer, the angle under which the grooves on the side of the optical fiber can be determined.

However, such an optical fiber can also be used to capture the light from a concentrated beam of light 50 that moves linearly 34 and to redirect it in the longitudinal direction of the optical fiber to an exit, as illustrated in Figure 3a .

As already mentioned in these publications, a slotted optical fiber of arbitrary length can be obtained by coupling separate segments (see SPIE article, chapter 2, last paragraph). Therefore, an optical fiber can be made that extends over the entire width W of the product flow. Therefore, for every 55

Immediate position of the concentrated light beam, such a slotted optical fiber can capture, in the absence of products, the uninterrupted light 34 and redirect it to a detector 40.

Because the angle formed by the uninterrupted light and the optical fiber used as the background element 5 in Figure 3a, depends on the position according to the width W, it may be necessary to vary the orientation of the grooves along the fiber optic length After all, the uninterrupted light beam 34 will enter the fiber substantially perpendicularly in the middle of the product flow, if the inspection unit 10 is arranged symmetrically in relation to the width of the product flow. However, moving to the edges of the product flow, the uninterrupted light beam 34 strikes the optical fiber 5 at a certain angle.

This can be avoided in two ways. As already mentioned, the optical fiber can consist of coupled segments. In each segment of the optical fiber, the orientation of the slots may be the same. This can lead to efficient production of these segments. Each segment of the optical fiber 5 can be placed under another angle, depending on the position according to the width of the product flow. A segment in the middle of the product flow will be placed in a substantially parallel position in relation to the product flow, while the segments at the edges of the product flow are placed at an angle in relation to the product flow, corresponding to the angle formed. by the light beam 34 and the flow of products. Due to this variable orientation 15 of the segments along the width of the product flow, the uninterrupted light 34 will always impact the grooves substantially at the same angle and will be captured and redirected in the same way. Preferably, the segments are placed in an arc described by the concentrated light beam 45 when the product flow is scanned.

As already mentioned, the optical fiber can consist of coupled segments. In each segment of the fiber optic 20, the orientation of the grooves can be changed. After all, the orientation of the slots can be chosen separately for each segment by consistently establishing the corresponding relative orientation of the printer. Therefore, the orientation of the grooves of a segment can be adapted according to their position in relation to the width of the product flow and the angle formed by the beam of the incident light 34 and the product flow at that point. In this embodiment of Figure 3a, all segments of the optical fiber are placed substantially parallel to the product flow. The orientation of the grooves of a segment, however, will depend on the position along the width of the product flow according to the angle formed by the uninterrupted beam of incident light 34 and this position. In this way, the uninterrupted light 34 will each time form substantially the same angle with the slots. The light will also be captured and redirected in the same way. 30

An alternative as illustrated in Figure 3b, consists in the application of the background element 5 as a plurality of small detectors 40 'that convert the incident light 34 into an electrical signal 39. In the embodiment illustrated in Figure 3b, the element Background can be constructed as a linear matrix or line of light sensitive elements, such as photodiodes or Photomultiplier Tube (PMT) or other elements known to those skilled in the art. Because such a linear matrix is generally constructed as a line of light-sensitive elements separated 40 ', that is, separated from each other, it is possible that an uninterrupted beam of light 34 does not strike a light-sensitive element 40 '. The erroneous conclusion could be reached that a product 2, 3 was present in the flow of products that obstructs the concentrated beam of light. To avoid the detection of discontinuous light, a limited light scattering effect could be provided when an uninterrupted beam of light reaches the bottom element 5. A coating can be applied on the bottom element that disperses the incident light, 40 when the same background element 5 is being scanned. This light scattering coating can be, for example, a milky plastic layer or a glass plate. With the use of this limited light scattering, it can be ensured that two or more nearby light-sensitive elements 40 'are illuminated, even if the uninterrupted beam of light 34 reaches the bottom element 5 between two such sensitive elements to the light 40 '. After all, it is important to know if an uninterrupted beam of light 34 reaches the background element 5, instead of the position in which the background element 5 is illuminated. This position can be deduced from the immediate known position of the scanning light beam and correlating the time period of the signal 47, coming from the background element 5, with the time period of the moving light beam.

The signal at the output of the detection background element 5, as illustrated among others in Figures 3a and 3b, can also be processed. The electrical (Figure 3b) or optical (Figure 3a) signal at the output of such a background element of detection 50 can be filtered to retain only the signal components from the uninterrupted light beams 34, while the ambient light It is leaking. The signal components usually have a higher frequency than that of the signal components from the backlight. A DC filter or high pass filter may generally be sufficient to allow only the desired high frequency signal components, characteristic of the presence of a product passage 2, 3, for subsequent signal processing, such as It is explained in this description.

Figure 4 schematically shows a detection unit 44, impacted by the incident light or light cone 46, 47, 48, and which subsequently converts said light, or a particular area thereof, into an electrical signal by means of a converter optical-electrical 38. This electrical signal 39 is provided as input to a unit of

processing 41, which by means of an analytical method generates a control signal 42 that controls a rejection system 11.

According to the invention, the optical filters 36 can be used to make the detection unit 44 sensitive to a specific wavelength by placing this filter 36 in operative communication with the aforementioned optical-electrical converter 38. 5

According to the invention, a space filter 37 can be used to block or pass through certain areas of the return light 46, 47, 48. For example, a space filter 36 can be used, which only lets the scattered light. Such space filters are described in US patents US 4634881 and US 4723659.

In a preferred embodiment, the spatial filter 36 consists of a diaphragm that is placed just in front of the optical-electric converter 38.

As schematically illustrated in Figure 5, more detection units 44, 44 'according to the invention can be arranged.

In a preferred embodiment, each detection unit 44, 44 'uses a different combination of optical filters 36 and space filters 37. Because of this, each detection unit 44, 44' is sensitive to a specific area of the return light 15 46, 47, 48 which has a specific wavelength. The output signals 39, 39 'are representative of a specific area of the return light 46, 47, 48 at a specific wavelength.

The first detection unit 44 generates a first electrical signal 39 with a level determined by the aforementioned optical and spatial filters chosen for that detection unit. The second detection unit 44 'generates a second electrical signal 39' with a level determined by the aforementioned optical and spatial filters 20 chosen for that detection unit. The detection units 44, 44 'are in operational communication with the processing unit 41 via signals 39, 39'.

The processing unit 41 will make a selection between the scanned products 2, 3 and the background element 5, depending on the return light 46, 47, 48, more specifically based on the electrical signals 39, 39 '.

In a preferred embodiment according to the invention, the processing unit 41 is a digital processing platform based on Field Programmable Door Matrices or microprocessors. The processing unit 41, however, could also consist of analog amplifier circuits or a combination of analog and digital components, as is known to those skilled in the art.

An apparatus described in the preceding paragraphs by combining a series of the characteristics described above is an apparatus 14 for classifying a flow of products 2, 3, comprising:

 a supply system 1 to supply the flow of products 2, 3 to a scanning area 28;

 a background element 5 located behind the scanning area 28;

 at least one inspection unit 9, 10 comprising:

to. a scanning unit 43 to illuminate the product flow along the entire width W of the product flow, and 35

b. a detection unit 44 for capturing an area of light returned by products 2, 3;

 a rejection system 11,

wherein said background element 5 consists of means for capturing an incident light 34 and for directing it to detection means 40, 40 ′ which are configured to convert said light into an electrical signal 39.

The detection unit 44 may consist of one or more separate units 44/44 ', as shown in Figures 40 4 and 5, arranged to receive light beams 46, 47, 48 reflected by products 2, 3 and the element background 5. The received light can be converted into electrical signals and further processed according to the methods described elsewhere in this description.

In the apparatus and method according to the invention, the detection unit 44 is a camera, configured to capture images of the falling products as they pass through the scanning area 28 in front of the bottom element 45. That is, the camera captures an image of the scanning area 28. In this embodiment, the background element 5 consists of means for capturing the incident light from the scanning unit and directing it to a detector 40, or the background element consists of a set of detectors 40 'configured to convert said light into an electrical signal (that is, the background element is produced in accordance with Figure 3a or 3b).

The camera can be sensitive (i.e. capable of detecting) to reflected laser light beams 46, 47, 48, reflected by the products and the background element. In this case the camera acts as a detection unit that is completely equivalent to the detection unit 44 shown in Figures 4 and 5. According to a more preferred embodiment, the scanned laser light 46, 47, 48 which is reflected for products 2, 3 and the background element is not used directly to make the selection. In this case, the camera is equipped with filters that make the camera insensitive to said reflected laser light. Instead, light from the environment or from an independent light source (that is, independent of a scanning medium), for example, a row of LEDs that emit pulsed LED lighting, is reflected in the products of the scanning area and captured by the camera. In another alternative, the camera is sensitive both to the reflected laser light 46, 47, 48 and to the light coming from the light source or the environment reflected by the products and the background. The independent light source, for example row 10 of LEDS, can be included in the apparatus according to the invention.

The camera can be a camera with a focal plane assembly, for example a solid state camera or a CCD camera, mounted properly with respect to the flow of products. A possible camera configuration suitable for an apparatus according to this embodiment is shown in WO2008 / 116924A2, which is incorporated herein by reference. The difference between an apparatus according to the last embodiment and the apparatus of WO2008 / 116924A2 is that the background element in an apparatus of the invention consists of means to capture incident light from the scanning unit and to direct it to a detection means 40, 40 'which is configured to convert said light into an electrical signal 39. The detection means is a detector 40 arranged to receive light 56 from a grooved fiber, as in Figure 3a, or is a plurality of detectors 40 'arranged along the surface of the bottom element 5, as in Figure 3b. According to one embodiment, an apparatus equipped with a camera then comprises means for producing a location signal that is indicative of the location of the products along the width W of the scanning zone, wherein said location signal it is based on the detected laser light that falls on the background element and thus passes through products 2, 3, that is, light that is not blocked by products 2, 3. The method for deriving such a location signal It is further described in this description in relation to Figure 13. The processing of said location signal together with the captured camera image allows an easy selection process, by checking whether electrical signals related to products 2, 3 and obtained by processing the camera image cross or not a predefined threshold.

The method according to an embodiment of the invention, as illustrated in Figure 6, is that the light 45 having at least one wavelength is sent to an inspection zone 28. This zone 28 is scanned 30 and when the aforementioned light 45 hits the product 2, 3 or the background element 5, the emission 46, 47, 48 is captured by at least two detection units 44, 44 '.

In the processing unit 41, the incoming signals 39, 39 'can be combined into new signals A, B according to the formula:

      A = n39 + m39 ', 35

B = p39 + q39 '

where n, m, p, q are real numbers and 39, 39 'are said input signals.

In an advantageous embodiment of the invention, the factors m and p are equal to zero. This means, in principle, that no combination is made. In this case, the detection unit 44 generates the signal A, and the detection unit 44 'generates the signal B. 40

By way of example, without limiting its scope, a detection unit 44 is considered to have an optical filter 36 arranged in the light spectrum between 690 and 740 nanometers, more specifically, the fluorescence spectrum of the product 2 which Contains chlorophyll when illuminated between 540 and 680 nanometers. Signal A shows a possible path of such a configuration in which a peak 16 is noticeable at the location of said product 2 containing chlorophyll. Four. Five

The problem occurs when the signal level in zone 17, due to the products to be rejected 3, does not show a noticeable difference with the background signal 15. In that case, it is not directly possible to make a distinction between the products to be rejected 3 on the one hand, and the bottom element 5 together with the products to accept 2, on the other hand. Although this distinction must be made because only products to be rejected 3 are allowed to originate a rejection action through a rejection system 11. 50

In Figure 6, signal A illustrates a path in which an emission phenomenon, which can only be attributed to the product to be accepted, is measurable. However, the selection of such a signal is not possible due to the problem mentioned above.

The signal B, as illustrated in Figure 6, is shown as a function of the width W of the scan zone 28. The zones 20, 21 of the signal are the result of the emission that occurs in products 2, 3 , in particular as 55

consequence of the issuance of both the products to be accepted 2 and the products to be rejected 3. Zone 19 is the result of the issuance of the fund element 5.

In signal B, a zone 49 is defined in which the background signal 19 is located. All zones 19 that are located within this zone 49 are labeled by the processing unit 41 as coming from the background element 5 To be precise, zone 49 is determined by a maximum threshold value tmax and a minimum threshold value 5 min. In an advantageous embodiment, these threshold values tmax, tmin can be adjusted by a user.

In order to better represent the following steps in the method according to the present invention, a Boolean signal C is introduced, whereby the value 0 is adopted at the locations of the background signal 54 and the value 1 is adopted in locations 18 outside of zone 49 of signal B.

The present invention, however, does not exclude that, in an alternative method, the Boolean signal C is effectively generated or directly available in the processing unit 41, for example in the case of an embodiment as described in the figures. 3a and 3b, in which the background element 5 generates a signal 39 in accordance with the path of the signal B, which can be transformed in the manner described above in said signal C. However, it must be observed that in this case, the selection between products 2, 3 and the bottom element 5, can be carried out using a threshold value tmin only because the peaks 20, 21 in which 15 products 2, 3 are located, are all on the same side of background signal 19, all below (or all above) of background signal 19 to be precise.

In a possible next step, according to the invention, a new signal D is generated by the formula:

D = AC + s (C⊕ 1),

where ⊕ is defined as the addition of module - 2. 20

Thus, the new signal D is an exact copy of the above-mentioned signal A in places 20, 21 of product 2, 3. In the places where the background element 5 is observed, a new value s is established. , so that the background signal 22 is clearly separated from the products to be rejected 3.

In the example described above, the signal path A was interpreted as coming from the emission peak of a chlorophyll-containing product 2, against a non-fluorescent background. By generating the signal D, a distinction can be made between the background element 5 together with the products 2 containing chlorophyll, on the one hand, and the products to be rejected 3, on the other hand, so it is possible to eliminate Automatically these latest products.

In a next preferred step, the signal D is further filtered, for example, by a low pass filter, generating a new signal E. The low pass filter is constructed according to known principles, for example 30 by a digital multitouch filter. FIR. The cutoff frequency will be chosen so that the high frequency transitions at the edges of the zones 20, 21 in the D signal are sufficiently flattened, without losing the actual signal content. In this way, the zones 24, 25, in which the products 2, 3, and the zone 23, in which the bottom element 5 is located, are obtained.

In an alternative form, the filtering is carried out by means of an adaptive filter that is tuned to apply a filtrate mainly to only such transitions. In that case, the cutoff frequency chosen may be much lower.

The present invention is not limited to the use of low pass filters or adaptive filters to flatten said transitions. To be precise, all methods for performing such flattening are within the scope of this invention. 40

Depending on whether the signal E crosses a specific threshold value g or not, an automatic detection can be carried out in places 25, where the products to be rejected are 3.

Instead of generating signals D and E, according to the invention, automatic detection can be performed in those places 17 where the products to be rejected are found, depending on whether signal A crosses a threshold value g, or no, simply by analyzing said crossing in those areas where products 2, 3 of 45 are found according to signal C.

The signals A, A ', B, C, D, E are synchronized with each other. After all, these signals originate from concentrated light beams 45, scanning the flow of products in temporary motion. Therefore, each immediate value of one of these signals can be correlated with the corresponding immediate value of the other signals. This synchronization allows the application of a signal to another signal or the combination of both, and at that moment, the signals arrive from the same scanning position. In Figures 6a-e, the signals A (tA), B (tB), C (tC), D (tD) and E (tE), where tA, tB, tC, tD, tE represent the time dependence of these signals, it

they can correlate with each other, since tA = tB = tC = tD = tE since all signals are obtained by a beam of light 45 that travels back and forth temporarily.

To actually eliminate the products 3, the air valves 11 are opened so that, in the places where these products 3 were detected, each such product 3 will be blown out of the product flow.

The operation of the rejection system 11 is carried out by comparing the control signals 5 generated by the signals D, E, with one or more threshold values gi. These control signals only contain information about the product to be rejected 3, if it has to be removed, or the product 2 to be accepted, if it has to be retained.

This method according to the present invention, of course, is not limited to the use of two detection units. In the case of more than two detection units, the corresponding signals 42 can, for example, be combined algebraically with the aforementioned signals A and B.

In an advantageous embodiment of the present invention, zone 49, in which the background signal is located, is defined in several signals B, to be precise in all output signals 39 of all the detection units present 44. The Final Boolean signal C that determines where the bottom 19 is located and where the products 2, 3 are found, is obtained by performing a logical operation - OR on all the separate signals C that are obtained according to the above-mentioned method.

In an alternative embodiment, the output signals A, A 'of two detection units 44, 44' are combined in a two-dimensional graph. Both signals A, A 'originate from the same position in the product flow, although they may differ in one or more signal parameters, so that different optical properties of the position scanned at that time can be analyzed. These signals A, A 'can be obtained by filtering them out of the same light signal detected using spatial and / or frequency filters. Each point in this graph corresponds to a specific intensity level according to the trajectory of the first signal A, combined with a specific intensity level according to the trajectory of the second signal A 'at a specific moment in time, or , in other words, for a known immediate position of the concentrated scan beam 45. Specific combinations of the signal can occur several times if products 2, 3 with the same optical properties are being scanned or each time the element Background 5 is being scanned. By maintaining these statistics, a two-dimensional histogram can be created. In addition, a color can be attributed to each value of the histogram. By assigning, for example, blue to the lowest value and gradually moving to red for increasing values, a two-dimensional intensity map 55 can be created. In this map 55 contours of equal intensity can be extracted. As shown in Figure 7a, areas in which 30 products with similar optical properties can be determined can be determined.

Each point in the two-dimensional diagram of Figure 7a is characterized by the intensity levels of the respective signals A, A 'and by the corresponding value of the location signal C. Depending on the value of this location signal, the levels The signal of a point in the histogram corresponds to those of a product 2, 3 or the background element 5. Based on the information of the signal C for the combination of individual signals A, A ', it can be determined whether a Product is affected or not.

In this way, in this perception, zone 51 is defined, in which the products to be accepted are 2, and zone 52, in which the products to be rejected are found 3. Zone 50, determined by the values respective threshold tmax, tmin and t'max, t'min, is represented by a square in said intensity map 55 containing the background as illustrated in Figure 7a. However, the signal combinations of a product to be rejected 3, 40 can be located between the threshold values T'max ', T'min and Tmax', Tmin containing the points corresponding to the positions scanned in the background element 5. Figure 6b illustrates how the signal level of the background signal 47 is within a band 49, limited by one or both threshold values Tmax, Tmin. When the background element 5 shows no variation in the corresponding signal 47, the background level may be a single line in Figure 6b and a single point in Figure 7a. In a realistic embodiment, this background level can vary in such a way that a corresponding set of points is obtained within the rectangle T'max ’, T'min and Tmax, Tmin defining the band 49 in Figure 7a. As shown in Figure 7a, some points belonging to a product to be rejected 3, may have signal levels that cannot be distinguished from the background element 5. This is illustrated in Figure 7a by overlapping between the band 50 and zone 52. This is also illustrated in Figure 6a in which the signal in position 17 of a product to be rejected 3 cannot be distinguished from the signal in position 15 from the bottom element 5.

In a graphical user interface, by means of a simple operation, as illustrated in Figure 7b, for example by a dragging movement, the square 50 can be moved to another location with coordinates n ', n, such that the area 52 which coincides with the products to be rejected 3, can be isolated by a separation plane g '. Based on the corresponding value of the positioning signal, it can be defined whether each point refers to a product 2, 3 or to the background element 5, regardless of the corresponding levels of the signals A, A '. As explained above and also illustrated in Figures 12a-d, this positioning signal C can be obtained from one or more observed light signals A, B whereby

You can use this positioning signal C to indicate, respectively, within this or other light signals, the position of the bottom 5 or of the products 2, 3.

In a very advantageous embodiment, zones 50, 51 and 52 can be calculated automatically by means of known grouping algorithms, for example means K.

The invention is not limited to presentations of one and two dimensions, but can be easily extended to 5 presentations of three and more dimensions, by projections of one, two or three dimensions in those cases.

Hereinafter, a detailed description is provided, as shown in Figure 8, of a possible practical construction of an apparatus 26 for performing the aforementioned method.

Figure 8 shows a complete sorting apparatus 26 in a perspective view. This apparatus consists of a delivery system 1, more specifically a vibrating table that carries a flow of products 2, 3 in a certain direction 27 through the sorting apparatus. During its free fall, the product is also guided by a free fall plate 4.

The structure 26 is further equipped with 2 inspection devices 9, 10. These inspection devices 9, 10 inspect an inspection area 28 by means of a concentrated beam of light that sweeps through the entire width W of the product flow. In the absence of products, a background element is scanned so that, according to the invention, it can have the optical characteristics of the products to be rejected.

The products to be rejected 3 are blown out of the product flow through the air valves. Accepted products 2 are guided through an axis 53 towards possible additional production stages.

During the linear scan of the product flow and the background element 5, signals from this background element 5 and the product flow are detected, in particular the good products 2 and the bad products 3. By 20 the adjustment of the filters of frequency and space control signals can be generated from the detected signals that allow classifying the products supplied according to previously established criteria using the rejection system 11. Depending on the chosen background element 5, the background signal 47 will have a higher or lower intensity value: a white background provides a higher value, a black background provides a lower value, as indicated in Figure 9b. The signals from the good products 46 and the bad products 48 overlap this background signal 47. Each product 2, 3, 3 'provides a corresponding signal peak 46, 48, 48', each characterized by a certain Pulse width and pulse height or signal level. Figure 9b schematically reflects this combined signal B, obtained from a scan in which two good products 2 and two bad products 3 were scanned as in figure 9a. The desired products 2 must be retained in the product flow, while the unwanted products 3, 3 'must be rejected by the rejection system 30 11. When neither a good product 2 nor a bad product 3, 3 is scanned ', rejection system 11 must remain inactive. Although at that time no product to be rejected is present in the system, unnecessary activation of the rejection system 11 could cause an undesired disturbance of the product flow. The reference level 70 in this combined signal B is located at the level of the background signal 47. In the current state of the art, based on this combination of signals represented in Figure 9c, the products 35 could be identified bad 3 by determining a first "negative" threshold value g1. These bad products 3 that give a signal peak 48 that exceeds the first threshold value g1 can be removed from the product flow. This threshold value is specified as negative throughout this description, since it is located below the reference level 70 in the reference system of Figures 9 a-e. Since the background signal 47 constitutes the reference 70 to determine the signal peak 48 and the threshold value g1, the level of this signal peak 48 always exceeds this first threshold value g1, as indicated by the area scratched in the figure 9c. Good products 2 have a signal peak 46 that does not exceed the first threshold value g1. As long as the bad products 3 have a signal peak 48 that remains below the signal peaks 46 of the good products and exceeds the first threshold value g1, the good products 2 can be differentiated from the bad products 3 based on these signals 46, 48, which ultimately allow products to be classified into a flow of good products 13 and a flow of bad products 45 12.

In some cases, however, it is possible for a bad product 3 'to generate a signal 48' from which the peak is smaller than the signal 46 from a good product 2. This issue has already been clarified in the illustrated embodiment in figure 6. The selection method presented in figure 9c does not allow distinguishing said bad products 3 'from the product flow as well. After all, when the negative threshold value g1 is shifted to the level of the reference value 70 of the background signal 47 in a way that the signal peak of the bad product 3 'exceeds the threshold value g1, the good products 2 are will remove too. The corresponding signals 46 have, after all, a maximum value that is greater than that of these bad products 3 '.

Such bad products 3 'could be identified by defining a second negative threshold value g2. The second threshold value g2 is chosen such that the signal peaks 46 of the good products 2 exceed 55 this second threshold value g2, while the signal peaks 48 'of such bad products 3' would not exceed the second threshold value g2 , as indicated by the striped area in Figure 9d. The signals that are located by

Below this threshold value g2 then match the products to be rejected 3 'of the product flow. Not only the signal peaks of the good products 2, but also the signal peaks 48 of the other bad products 3 exceed the second threshold value g2 and beyond the first threshold value g1 as described above. A problem that can occur with respect to the choice of threshold value g2 in Figure 9d is that all signals, also those of good products 2, pass partially below and partially above 5 of threshold value g2 and could thus be misinterpreted as coming from a bad product. This is also illustrated in Figure 9f, in which the signal peaks of a product to be rejected 3 'are located in a band limited by two threshold values g2, g2', which contains the signal peaks of a product type to reject. Here too the signal peaks of the good products 2 will pass through such a band and will be mistakenly considered as coming from a product to be rejected 3 '. 10

By means of a suitable choice of the threshold values g1 and g2, the bad products 3, 3 ', characterized respectively by a high signal peak 48 and can be identified from the combined signal, as indicated in Figure 9e. a low signal peak 48 '. Only signals 46 within the band consisting of two threshold values g1-g2 are considered as coming from a good product 2. These threshold values could also be described, respectively, as an upper limit g1 and a lower limit g2, jointly limiting a 15 signal band location of the signal peaks in which the rejection system 11 cannot react.

The selection method in the current state of the art as presented in Figures 9 a-f, however, does not work without errors. Since the background signal 47 is to be used as a reference to determine the signal peaks 46, 48, 48 ’and the threshold values g1 and g2, this background signal 47 is always above the second negative limit g2. As with bad products 3 ’, the rejection system 11 will react in the presence of the background element 5. Consequently, the rejection system 11 will react to:

- product signals 48 with a signal peak that exceeds the first negative threshold value g1;

- product signals 48 ’with a peak that does not exceed the second negative threshold value g2; Y

- background signals 47 which, by definition, do not exceed the second negative threshold value g2.

Good products are in fact located in the band between a first negative threshold value g1 and a second negative threshold value g2. Good products can be classified using this procedure by proper selection of values. The problem with the fund, however, is not solved: this, as illustrated in Figures 9 a-e, is considered to be a bad product.

One way to solve this problem is by physically constructing the background element 5 in a way that provides a signal 47 that is comparable, for the detected optical parameter, with the signal 46 from 30 of a good product 2 and therefore is within the band consisting of two threshold values g1, g2. As mentioned above, it is, however, difficult to construct a background element 5 in such a way that for the measured optical parameter (s), it not only resembles the real good product 2 well enough, but can maintain this similarity of a specific product 2 to a sufficient extent and for a considerable time. In addition, the problem remains that a corresponding bottom element 5 has to be installed for each product 2.

Another way to solve the problem is to shift the reference level 70 of the combined signal B using signal processing techniques, to a new value 71 that preferably coincides with the signal level 46 of a good product 2. The signal peaks 48 , 48 'of the bad products 3, 3' are being referenced in the processed signal D to the new reference level 71. To clarify the invention, this new reference 71 is adjusted to the signal level 46 of a good product 2. This It is illustrated in Figure 10c. When initially the reference level 70 is in the background signal 47, as indicated by the signal B in Figure 10b, the reference level 71 for the new signal D is substantially equal to the signal level 46 of the good products 2. Through this change in the reference level of 71 -> 71, which results from signal processing, the corresponding signal peaks 48 of some bad products 3 will still extend from the new reference level 71 down. These are 45 unwanted products 3 that have, in the original reference frame of signal B as illustrated in Figure 10b, a signal peak 48 that exceeds this peak 46 of the desired products 2. These unwanted products 3 they still give rise to a negative signal peak even in the new reference frame 71 of the signal D. For any other unwanted products 3 ', the corresponding signal peaks 48' will now extend from the new reference level 71 upwards . These are the unwanted products that, in the original reference frame 50 illustrated in Figure 10b, had a signal peak 48 'that did not exceed these peaks 46 of the desired products 2. Instead of providing a peak of negative signal, these unwanted products 3 'provide a positive signal peak 48' in the new reference frame 71. By changing the level 71 with respect to which the signals of different products were being referenced, it is now as if the signal of given background 47 with level 70 will be removed from the detected signal B and replaced by a new signal with a level 71 in signal D. 55 With this replacement of reference level 71 a new signal D is obtained, as indicated in the Figure 10c, which now contains information on products 2, 3, 3 'such that these products can be differentiated from each other without the problems present in the prior art.

Now good products 2 can be identified by defining an appropriate threshold value or values for the new signal D, as shown in Figure 10d, while it is necessary to distinguish between 2 types of bad products 3, 3 ’. As for the product signals 46, 48, 48 ', these threshold values refer to the new reference level 71, which, in this example, has been chosen to be substantially equal to the signal level 46 of the good products 2. A first negative threshold value g1 has been chosen such that the 5 unwanted products 3 provide a signal peak 48 that exceeds this first threshold value g1. These products 3 can be rejected from the product flow by activating the rejection system 11. A second now positive threshold value g2 is chosen such that the unwanted products 3 'provide a signal peak 48' that exceeds the second value g2 threshold. This threshold value is indicated as positive because in the given reference frame, this threshold value g2 is located above the reference level 71. These 10 products 3 'can be rejected from the product flow by activating the rejection system 11. Since the signal level 70 of the background signal 47 according to the background element 5 moves to the new reference level 71, in this case, when the signal of the good products 2 is matched, the Variations in the signal level of the background signal 47 will be within both threshold values g1, g2. Therefore the rejection system is not activated. fifteen

The signal levels 46 of the good products 2 may show small variations, provided that these variations are around the reference level 71 within the threshold values g1, g2. The rejection system is not activated in that case.

By properly choosing the threshold values g1 and g2, bad products 3, 3 'can be identified from a combined signal as indicated in Figure 10d, which are characterized by a high signal peak 48 and a signal peak 20 under 48 ', respectively. All signals 48, 48 ′ outside the band defined by both threshold values g1 - g2 compared to reference level 71 are considered to be derived from a bad product 3, 3 '. Although in the embodiment illustrated in Figures 10 a-d a signal with 2 types of bad products 3, 3 'is shown, the invention is not limited to such signals. Also in the B signals in which only one type of bad product is produced, either 3 or 3 ’, the signal processing technique can be applied as illustrated in the different embodiments of the invention. This is illustrated in Figures 15 a-c and in Figure 6, in which only an undesired product of type 3 'is produced, that is, with a signal peak 48' compared to the original reference level 70 which is less than the signal peak 46 of the good product 2. By changing the reference level 71 this signal peak 48 'can be distinguished without a doubt from the signal peak 46 without the problems of the current state of the art, such as unwanted reaction of the rejection system 11 to a background signal 47 or to a signal peak 46 that would exceed the lower limit 30 g2 when the original reference frame 70 is still used for further signal processing.

By changing the reference level 70 to a new suitable value 71 one can better distinguish between the different products 2, 3 and / or 3 'and avoid unnecessary activation of the rejection system 11 either when the background signal 47 is detected and / or when a signal 46 passes through a lower limit g2.

Because the reference level 70 of the background 47 is shifted to a new level 71 in the embodiments according to the invention using signal processing techniques, a background signal 47 is taken which differs, preferably considerably, of any product signal 46, 48, 48 '. When the background reference level 47 is shifted by signal processing, preferably at the signal level 46 of a good product 2, the exact value of this background level 47 is not important, as long as the background element 5 provides a signal 47 that differs, preferably considerably, from signals 46, 48, 48 'of 40 any product 2, 3, 3'. This choice of the background signal 47 makes it possible to clearly distinguish between the original reference level 70 and the product signals 46, 48, 48 ’and, as such, refer these product signals to the new level 71.

In the prior art, there are different ways of determining the level 70 of this background signal 47. The inspection configuration or configurations 9, 10 can be allowed to function without supplying any product. The selected signal 45 B substantially coincides with the signal 47 of the background element 5. Known products 2, 3, and / or 3 'can also be inserted in the known positions in the scanning beam of the inspection unit 10. The signal immediate that is being detected, that is, at the given time or moments, it coincides with a given immediate position of the scanning light beam and therefore with a given position x (mm) in the product flow. In this way, the different signals 47, 46, 48 and / or 48 ’can be identified and correlated in signal B with the bottom element 5 and with the products transported 2, 3 and / or 3 '. Instead of the static tuning procedure indicated above, you can also proceed in a more flexible way. As illustrated in Figure 1, a classification system can work so that the product flow is scanned and B signals are detected. Initially, all products are accepted or rejected. By gradually adjusting the threshold values g1 and / or g2, a selection is made within the product flow. The threshold values may continue to be adjusted until clearly only unwanted products 3, 3 ’are rejected. Figure 7 illustrates such dynamic specification of the different signal levels and the corresponding setting of the threshold value or values.

As indicated in the previous paragraphs, it is possible to redefine 71 the reference level 70 of the detected signal B by signal processing, so that a new signal D is obtained that allows products 2, 3 and / or 3 'are distinguished from each other without the problems present in the prior art. 60

Preferably, the signal 46 from the good products 2 and the signal 47 from the background element 5 are displaced at a substantially equal signal level 71, at least by the detected optical parameter.

In order to determine which zone of the signal B corresponds to the background signal 47 and, therefore, in which zone the signal level 70 must be replaced by a new suitable reference level 71 for the product signals 46, 48 and 48 ', it can be determined in the original B signal which zones 46, 48 and 48' indicate the presence of the products 5 2, 3, 3 'in the linear scan and which zones 47 indicate the absence of a product or, in other words, the presence of the background element 5 in the linear scan.

There are different embodiments to obtain a C signal based on one or more detected signals, in which the pulses are indicative of the location of a product 2, 3, 3 ’. This location signal C therefore contains information on the desired products 2 as well as the unwanted products 3, 3 ’. 10

Figures 11 a-d illustrate a first embodiment. In the detected signal B, zones corresponding to the background signal 47 can be distinguished from the zones corresponding to product signals 46, 48 and 48 ’. As stipulated above, because the signal 47 from the background element 5 is chosen in such a way that it differs from the signals 46, 48 and 48 'of the products 2, 3, 3', one or more can be defined threshold values g3 and g4 in signal B, such that signal levels 46, 48 and 48 'of products 2, 3, 3' are 15 outside these threshold values, for example on one side of this value threshold g3 and signal level 47 of background element 5 is within these threshold values, for example, on the other side of this threshold value g3. Figure 6 illustrates the embodiment in which two threshold values g3 / tmin and g4 / tmax containing the background signal 47 have been determined. Figure 11b illustrates an embodiment in which a threshold value g3 is simply used to define the signal background 47 against product signals 46, 48 and 48 '. In this signal analysis, the signal peaks 46, 48 and 20 48 'and the threshold value g3 are determined with respect to the original level 70 from the background signal 47. Each peak 46, 48, 48' that exceeds this Threshold value g3 thus indicates the presence of a product 2, 3, 3 '. A signal C is obtained that has a pulse every time a product 2, 3, 3 ’is detected in the linear scan. In figure 11c, a signal C of this type is shown, so that the product pulses 46, 48 and 48 'are already inverted compared to the original signal B. This signal C can also be a binary signal as shown in figure 25 11d. When converting the analog signal into a digital signal, a pulse train C of pulses "1" and "0" is obtained, where "1" indicates the presence of a product 2, 3, 3 'and "0" indicates the absence of a product 2, 3, 3 '.

In this embodiment, the same detected signal B is first used to generate a product location signal C, after which, as shown above, this detected signal B is combined with the signal C to change the level reference, as illustrated in figures 10 a-d. 30

Figures 12 a-d provide a schematic representation of different embodiments of this signal processing process. In the embodiment illustrated in Figure 12a, the detected signal B and the threshold value g3 are compared in a signal processing unit 60 to generate the product location signal. This signal C is then combined in a signal processing unit 61 with the detected signal B in order to determine which areas of this detected signal B match the signal 47 from the background element 5. The reference level 70 of These background signals are then moved to the desired level 71, preferably the level 46 of the good products 2, which thus generates a signal D with an adjusted reference level 71 which makes it possible to distinguish the signal peaks of the good products 2 of these of the unwanted products 3, 3 'in the product flow. Due to the choice of threshold values g1 and g2, it is possible to distinguish between signals 48, 48 ’from unwanted elements in the product flow and signals from desired elements 2. 40

As explained in the previous sections, it is not necessary that signals C and B come from the same detected signals B. Figure 12b illustrates a signal processing process in which a signal C is obtained based on a first signal B. This signal C is then used to determine the position of signal peaks 46, 48, 48 'within the second signal B'. The location signal C is obtained based on one or more first detected signals B after which this location signal C is used to indicate the position of the products in one or more second detected signals B '. Both signals B, B 'are detected during the same linear scan because the return light from the product flow is converted through suitable space and / or frequency filters into different signals B, B'. Since these signals immediately correlated with each other, as mentioned above, the information of one signal can be combined or applied to another signal. In this way, the location of the product obtained can be used based on the first 50 signal B, to indicate in another signal B 'where the zones 16, 17 from the products 2, 3, 3' and the zones are located 15 of the background element 5.

As illustrated in Figures 12c and 12d, the location signal of product C can also originate from different signals Bi. By scanning the flow of products in different ways, for example with light beams 45 having different frequencies, or by analyzing the light returning from the flow of products in 55 different ways, for example by space filters and / or Frequently suitable, you can get a more complete picture of the product flow and prevent a product from remaining unnoticed. First, these signals Bi can be combined in a signal processing unit 62 and transform the combined signal B into a signal C as illustrated in Figure 12c. The signal processing unit 62 may

combine these detected Bi signals in any possible way: addition, subtraction, multiplication ..., as shown in Figure 12c. The different Bi signals can also be converted first into the corresponding product location signals Ci, which are subsequently combined in a signal processing unit 62 into the location signal of the desired product C, as shown in Figure 12d . Here, the signal processing unit 62 will combine the different independent positioning signals Ci by means of an "OR" function so that no product location information is lost.

Figures 13 a-c illustrate another embodiment for obtaining the product location signal C. In the embodiment illustrated in Figures 11 a-d, a signal 47 is generated during scanning by the background element 5. This background signal it may be a result of the reflections of the incident optical signal 34, or of the fluorescence of this background element 5 due to this exposure. Background signal 47 'was captured in this embodiment along with 10 signals 46, 48, 48' from products 2, 3, 3 ’. This alignment is also illustrated in Figure 2.

In the embodiment of Figures 13 a-c, however, the background element 5 does not re-emit any signal, but only detects the incidence of immediate light on this background element 5. The light 45 generated by the light source 29 , it is being blocked by products 2, 3, 3 'that are between the bottom element 5 and the light beam 45. As illustrated in Figure 13a and Figure 2, when the products are scanned, they will emit a light signal 15 46, 48, 48 ', for example, by reflection or by fluorescence. Only the light 34 that was not blocked by these products 2, 3, 3 'will impact the bottom element located later 5. When the bottom element 5 is provided with elements 57, 40 that allow the capture and detection of the incoming light 34, a signal C is obtained, which is indicative of the presence of products 2, 3, 3 'in the product flow for a given position of the scanning light beam, or in other words, at a specific time t ( s) or, in other words, for a specific position x (mm) according to the linear scan. After all, the position of the light beam 45 is known during scanning of the product flow, so it will not be a problem to correlate the temporal path of the signal 47 ', captured by the background element 5, with the temporal path of signals 46, 48, 48 ', coming from products 2, 3, 3' and therefore with the x (mm) position of a product 2, 3, 3 'in the product flow. Figure 13c illustrates the signal obtained C ', this time only showing peaks in which no product 2, 3, 3' 25 is present in the flow of scanned products. In Figure 13c, these peaks are labeled by the background reference 5, that is, the incidence of the light beam 34 on the background element 5. This analog signal C 'can also be transformed into a digital signal, therefore it obtains a pulse train C of signals of "1" and "0", in which "1" indicates the absence of a product 2, 3, 3 'and "0" indicates the presence of a product 2, 3, 3 '. The person skilled in the art will understand that this digital signal C ', if desired, can be easily transformed, using signal processing techniques, into a pulse train C, where "1" indicates the presence of a product 2, 3, 3 'and "0" indicate the absence of a product 2, 3, 3', so that a signal is obtained as shown in Figure 11d.

An advantage of the embodiment illustrated by Figures 13 a-c and in Figures 3a and 3b, is that the bottom element 5 can be placed at a greater distance d from the product flow. This prevents the bottom element 5 from being contaminated by the flow of products. In this configuration, the only essential question is whether for each immediate position of the moving beam of light, a signal is obtained in the background element 5 or not. Because a concentrated beam of light 45 is used, such as a laser beam, the light will not diverge substantially once the product flow has passed, even if the bottom element 5 is not being placed in direct proximity to the flow of scanned products. Consequently, the solid angle at which the concentrated light beam 34, for an immediate position of the scanning light beam 45, impacts the background element 5 is small enough to allow a precise distinction between the different positions of the moving beam of light in the linear scan.

Figure 14 provides a schematic presentation of this signal processing process, from the signal obtained, as illustrated in Figures 13 a-c. The product location signal C, from the background element 5, is combined 61 with the detected signal B to thereby determine which areas of this detected signal B correspond to a signal 47, from the background element 5. The level 70 of these background signals moves to the desired level 71, preferably level 46 of good products 2, thus generating a signal D with an adjusted reference level that allows distinguishing between the signal peaks of good products 2 and those of the unwanted elements 3, 3 'in the product flow. According to the choice of the 50 threshold values g1 and g2, the signals 48, 48 'of the unwanted elements in the product flow can be distinguished from the signals of the desired elements 2, as illustrated in Figures 10 c - d.

Figures 3a and 3b illustrate different embodiments of such a background element 5 that can detect uninterrupted beams of light 34.

As illustrated in Figures 12a-d and 14, the processing unit 41 comprises, according to the different embodiments, means 60, 62 for generating a location signal C based on one or more detected and converted signals B. In addition , this processing unit 41 comprises means 61 for generating, based on this location signal C, a signal D based on the same or other or other detected and converted signals B, whereby the background level (70) of The latter signals move to a new level (71) that allows a clearer and more effective distinction between good products (2) and bad products (3) in this signal D. 60

A method described in the preceding paragraphs through the combination of several of the characteristics described above is the method for classifying a flow of products 2, 3 into products to be accepted 2 and products to be rejected 3, comprising the steps of:

• move the products to be classified through a scanning area 28, supplied in a product flow that extends over a given width W and has a thickness of substantially a single layer of 5 products,

• in this scan zone 28, scan linearly across the width (W) of this product flow by one or more concentrated light beams 45, which illuminate, in the absence of products 2, 3, a bottom element 5 placed behind this product flow and extending over the width (W) thereof, whereby this light beam 45 produces light signals 46, 47, 48 in these scanned products 2, 3 and in this scanned background element 10 5,

• detect these light signals (B: 46, 47, 48) whereby these light signals are converted into electrical signals,

• generate one or more control signals based on these converted signals (B: 46, 47, 48), so that these control signals make it possible to make a selection of the scanned products to accept 2, on the one hand, and 15 scanned products to reject 3, on the other hand, and

• classify the flow of products 2, 3 using these one or more control signals,

in which

• the bottom element 5 consists of a means for capturing the incident light 34 and directing it towards detection means 40, 40 'which are configured to convert said light into an electrical signal 39, that is, the bottom element 5 can be an optical fiber with a grooved surface or a plurality of small detectors, and

• said step of generating one or more control signals comprises:

• generate a signal (C) that is indicative of the location 20, 21 of the scanned products 2, 3,

• generate said control signals depending on whether the light signals (B: 46, 48) produced in the scanned products 2, 3 cross, or not, a threshold in the areas where a product is present in accordance with said signal 25 (C) indicative of the location of the scanned products.

In the previous embodiment, the location signal is preferably obtained by detecting the light that falls directly on the background element and thereby passes through the products 2, 3, that is, the light that is not blocked by the products 2, 3, as described above in relation to Figure 13.

In the last embodiment, instead of generating signals D and E, an automatic detection is carried out in those places 30 where the products to be rejected are found, depending on whether the signal A crosses, or not, a threshold value , simply by analyzing said crossing in those areas where the products 2, 3 are located according to the location signal C.

An apparatus for classifying products according to the method of the last embodiment, comprises:

 a supply system 1, 4 that transports the products to be classified in the form of a product flow 35 that extends over a width (W) consisting of a single layer of products, in a certain direction 27;

 means 9, 10 for scanning the products to be classified 2, 3 across the width (W) of the product flow, in which these scanning means further comprise;

or means 29 for generating a concentrated beam of light 45 and directing it towards products 2, 3 through 40 optical means 30;

or means 44 for detecting the return light 46, 48 and converting it into an electrical signal;

 means 41 for generating control signals that allow a selection to be made among the scanned products 2, 3 based on said detected light 46, 48; Y

 means 11 for classifying the flow of products 2, 3 based on said selection by means of the said one or 45 more control signals,

wherein the sorting apparatus 14 further comprises:

 a background element 5 consisting of means for capturing the incident light 34 and directing it to detection means 40, 40 'which are configured to convert said light into an electrical signal 39, that is, the background element 5 may be an optical fiber with a grooved surface or a plurality of small 50 detectors, and

wherein the selection means 41 comprise:

 means for generating a location signal (C) that is indicative of the location 20, 21 of the scanned products 2, 3,

 means for generating one or more control signals depending on whether the light signals produced in the scanned products 2, 3 cross, or not, a threshold in the areas where a product is present in accordance with said signal of location (C) indicative of the location of the scanned products.

According to the invention, the method of obtaining the location signal illustrated in Figures 13a-13d can be combined with a classification method based on the analysis of a camera image of the products that are present in the area. Scan

According to the last embodiment, the method comprises the steps of:

• move the products to be classified through a scanning area 28, supplied in a product flow that extends over a given width W and has a thickness of substantially a single layer of products,

• in this scan zone 28, scan linearly across the width (W) of this product flow by one or more concentrated light beams 45, which illuminate, in the absence of products 2, 3, a background element 5 15 placed behind this product flow and extending over the width (W) thereof,

• capture a camera image of said scanning area, and deduce from said image, electrical signals representative of each product in said image,

• generate one or more control signals based on said image, whereby these control signals make it possible to make a selection of the products scanned to accept 2, on the one hand, and the products scanned to 20 reject 3, on the other hand , Y

• classify the flow of products 2, 3 using these one or more control signals,

in which

• the bottom element 5 consists of a means to capture the incident light 34 and direct it towards detection means 40, 40 ’which are configured to convert said light into an electrical signal 39, and 25

• said step of generating one or more control signals comprises:

• generating a signal (C) that is indicative of the location 20, 21 of the scanned products 2, 3, in which the generation of the location signal (C) comprises detecting and converting that part 34 of the scanning light beam concentrate 45, which passes through products 2, 3, that is, is not obstructed by products 2, 3, thus obtaining a signal that is indicative of the location of the scanned products 2, 3, as described with relation 30 to figures 13a-13c.

• generate said control signals depending on whether the electrical signals representative of said products and derived from said camera image cross, or not, a threshold in the areas where a product is present according to said signal (C) indicative of the location of the scanned products.

In the last embodiment, the image can be taken by any suitable camera system, such as for example a system described in WO2008 / 116924A2, which is incorporated herein by reference.

Claims (16)

1. Apparatus for classifying a product flow (2, 3), comprising:  a supply system (1) to supply the flow of products to a scanning area (28);  a background element (5) located behind the scanning area;  at least one inspection unit (9, 10) comprising: 5 or a scanning unit (43) to illuminate the product flow along the entire width W of the product flow, and or a detection unit (44) to capture an area of light returned by the products;  a rejection system (11); characterized in that said bottom element (5) consists of a means for capturing incident light and for directing it to 10 detection means (40, 40 ') that are configured to convert said light into an electrical signal, said apparatus further comprising a means for generating a location signal (C ') that is indicative of the location of the products, by detecting the immediate incidence of light on the background element (5), in which said detection unit (44) It is a camera arranged to capture an image of the scanning area (28). fifteen 2. Apparatus according to claim 1, wherein said camera is sensitive to the laser light beams (46, 47, 48) produced by said scanning unit and reflected by the products (2, 3) and the element of background (5). 3. An apparatus according to claim 1, wherein said camera is insensitive to the laser light rays (46, 47, 48) produced by said scanning unit and reflected by the products (2, 3) and the element of background (5), and in which said camera is configured to capture the light coming from the environment or produced by an independent light source 20, and reflected by the products (2, 3) and the background element (5). 4. Apparatus according to claim 1, wherein said chamber is both sensitive to: - laser light beams (46, 47, 48) produced by said scanning unit and reflected by the products (2, 3) and the background element (5), and - the light coming from the environment or produced by an independent light source and reflected by the products (2, 3) 25 and the background element (5). 5. Apparatus according to claim 3 or 4, wherein said individual light source is a row of LEDs that emit pulsed LED illumination. 6. Apparatus according to any of the preceding claims, wherein said camera is a focal plane matrix chamber. 30 7. Apparatus according to any of claims 1 to 6, wherein the background element is an optical fiber with a grooved surface to capture incoming light. 8. Apparatus according to any one of claims 1 to 6, wherein the background element is a plurality of detectors configured to convert the incident light into an electrical signal. 9. Method for classifying a flow of products in products to be accepted and in products to be rejected, comprising the 35 stages of: • move the products to be classified through a scanning area, supplied in a product flow that extends over a given width and has a thickness of substantially a single layer of products, • in this scan zone, scan linearly across the width of this product flow using one or more concentrated light beams, which illuminate, in the absence of products, a background element placed behind this product flow and that extends over the width of it, characterized in that the method further comprises: • capture a camera image of said scanning area, and deduce from said image, electrical signals representative of each product in said image, • generate one or more control signals based on said image, whereby these control signals allow a selection of among the scanned products to be accepted, on the one hand, and the scanned products to be rejected, on the other hand, and • classify the flow of products through these one or more control signals, further characterized by the fact that • the background element consists of a means to capture the incident light and direct it towards a detection means that is configured to convert said light into an electrical signal, and • said step of generating one or more control signals comprises: • generate a signal that is indicative of the location of the scanned products, in which the generation of the 5 location signal comprises detecting and converting that part of the concentrated scanning light beam, which is not obstructed by the products, thus obtaining a signal that is indicative of the location of the scanned products, • generate said control signals depending on whether the electrical signals representative of said products and derived from said camera image cross or not a threshold in the areas where a product is present in accordance with said signal indicative of the location of the scanned products. 10. Method according to claim 9, wherein the camera used to capture said image is sensitive to laser light beams (46, 47, 48) produced by said scanning unit and reflected by the products (2, 3) and the background element (5). 11. The method according to claim 9, wherein the camera used to capture said image is insensitive to laser light beams (46, 47, 48) produced by said scanning unit and reflected by the products (2, 3 ) and the background element (5), and in which said camera is configured to capture light from the environment or produced by an independent light source, and reflected by the products (2, 3) and the background element (5 ). 12. Method according to claim 9, wherein the camera used to capture said image is sensitive at a time to  - laser light beams (46, 47, 48) produced by said scanning unit and reflected by the products (2, 3) and the background element (5), and - the light coming from the environment or produced by an independent light source and reflected by the products (2, 3) and the background element (5). 25 13. A method according to claim 11 or 12, wherein said independent light source is a row of LEDs that emit pulsed LED illumination. 14. A method according to any of claims 9 to 13, wherein the camera used to capture said image is a focal plane matrix camera. 15. Method according to any of claims 9 to 14, wherein the bottom element is an optical fiber with a grooved surface to capture incoming light. 16. A method according to any of claims 9 to 14, wherein the background element is a plurality of detectors configured to convert the incident light into an electrical signal.