MXPA99003977A - Method for testing the reliability of a testing apparatus, specially an empty bottle inspecting device - Google Patents

Abstract

To test the reliability of a testing apparatus which checks multiple objects of the same type for one set feature, generates a feature signal for each object and checks the feature signal with respect to fulfillment of one first set condition, a test object is inserted in the testing apparatus after multiple objects and the feature signal of the test object is checked with respect to the fulfillment of a second condition, the second condition being a feature signal of the test object which corresponds to a previously inputted reference value. The reference value can be inputted by storing the feature signal of a test object as reference value.

Description

METHOD FOR VERIFYING THE RELIABILITY OF AN EXAMINATION DEVICE, IN PARTICULAR AN EMPTY BOTTLE INSPECTOR The invention relates to a method for verifying the reliability of an examination apparatus that inspects a large number of objects of the same type with respect to a characteristic, generates a characteristic signal for each object and verifies that the characteristic line satisfies a first condition, since a test object is conducted to the examining apparatus after a number of objects and it is verified that the characteristic signal of the test object satisfies a second condition . According to the state of the art, the procedure in the reliability checks for examination apparatus, for example those for empty beverage bottles known as empty bottle injectors, is that a row of test bottles is prepared so that each it contains a certain defect according to the defect recognition specification, that is, it does not satisfy a certain characteristic. A particular test bottle is prepared for each feature that is inspected. The batch of test bottles is then incorporated into the bottle flow at certain time intervals, for example every half hour, or after a certain number of bottles, by example 50,000 bottles. This operation can be automatic or manual. The test bottles are marked so that they can be instantly recognized as test bottles. The reliability test consists in verifying if the examination apparatus, for example the empty bottle inspector, can recognize that these test bottles are defective. In the method used hitherto the second condition is therefore complementary to the first condition, that is, the second condition is satisfied in the test bottles if the examination apparatus recognizes that the test bottle is defective, i.e. first condition is not satisfied. A record is kept in the examination apparatus. If the reliability test fails, that is, if one or more of the test bottles is not recognized as faulty (s), the test must be repeated. This "is to guarantee reliable operation, that is, the reliability of an examination device. This reliability test is unsatisfactory because it is only afterwards that it is recognized that an examination device is no longer operationally reliable. The reasons for the failure of an examination device are usually due to a dirty lens system or to the failure of individual components of the recognition electronics.

With the method according to the state of the art it was necessary to use a large number of test bottles to verify the reliability of, for example, an empty bottle inspector, each bottle having only one defect, for example a thread of defective closure or a foreign body in a singular recognition zone. Therefore, each type of defect and each recognition zone required its specifically prepared test bottle. If a test bottle had revealed several defects, the fact that this bottle was removed would not have ensured that all defects were recognized. - Accordingly, a set of test bottles is composed, for example, of about 10 to 15 bottles. The object of the invention is to create a method by which a deterioration in the operational functioning of an examination apparatus can be detected as soon as possible. According to the invention, this object is achieved by proceeding in such a way that the second condition is that the characteristic signal of the test object corresponds to a previously entered reference value. The idea on which the method according to the invention is based is to verify the reliability of the examining apparatus by means of a test object whose signal The characteristic is known with great precision when the examining apparatus functions correctly, and that it is entered into the examining apparatus as a reference value. Apart from unavoidable inaccuracies when inspecting the test object, the characteristic signal must correspond exactly to the reference value. It does not matter if the characteristic signal satisfies the first condition, that is, if the test object is free of defects. - For non-defective objects the characteristic signal can be located within a wider range or above or below a threshold value. The method according to the invention lends itself in particular to verify the reliability of inspectors of empty bottles. A procedure that is frequently used with inspectors of empty bottles when the absence of foreign bodies is verified is that an image of the object is scanned by points, for example by means of a CCD camera, generally in two directions that are at right angles to one another. of the other, and the brightness of each point of the image is confirmed and the clear-dark and dark-light transitions are recorded by comparing the brightness of adjacent image points. This type of transition always happens, for example, if the explorer passes over the edge of a body strange in an empty bottle. Even empty bottles that are free of foreign bodies have a certain number of brightness transitions, for example up to 100 transitions of brightness due to irregular areas in the wall of the container or the fluted at the edge of the base. A singular object has in this case "As free of foreign bodies up to 100 brightness transitions, that is to say that a characteristic signal of 100 still satisfies the first condition, during the correct operation, for example of an inspector of empty bottles, for the predominant majority of empty bottles, for example 90%, a characteristic signal is obtained which is a little below the number of 100 clear-dark transitions.If the sensitivity of the examination device of the examination apparatus drops due to dirt or other reasons, this tends to lead to a decrease in the number of clear-dark transitions that it recognizes per empty bottle.If the reference value of a test bottle is for example 95, the characteristic signal of the test bottle matches with a decline in the sensitivity of the recognition device, depending on the magnitude of the deviation from the reference value can be Choose several measures. For a deviation of, for example, 10% you can simply issue an alarm signal, while that for a deviation of 20% or more the examiner apparatus and the entire transportation apparatus can be stopped. A particularly advantageous version of the method according to the invention is obtained in conjunction with defect recognition methods in which not only the number of light-dark transitions are counted, but also the light-dark contrast of the transitions of brilliance. The image elements found that deviate from the background are divided into, for example, eight different kinds of brightness, or the clear-dark transitions are divided into, for example, eight different contrast groups, the number of transitions being light-dark is only counted according to this classification and is compared within each class with a particular threshold value. If the examination device of the examining device is dirty, a scattering of light appears in the lenses or protective glasses of the glass, which leads to a decrease in the light-dark contrast when covering a fogging effect in a vaporized manner. recognition device due to scattering of light. The reduction in light-dark contrast causes a shift of the light-dark transition within the classes of contrasts, in such a way that characteristic signals fall more frequently in classes with less contrast. When verifying the reliability of the examining apparatus, there is no longer a need to rely solely on comparing the number of clear-dark transitions, but a comparison is made with a threshold value in each of the contrast classes and, additionally, it can be take into account the distribution of characteristic signals through the individual contrast classes. Assuming that a test bottle has 100 clear-dark transitions in the order of 250 shades of gray, and two kinds of contrast are used of which the first class of contrast contains clear-dark transitions with less than 180 shades of gray and the second class of contrast contains clear-dark transitions with more than 180 shades of gray: if the lens system of the recognition device is slightly dirty, 100 clear-dark transitions are still recorded, but with reduced contrast, for example only 150 shades of gray. The characteristic signal set then indicates "100 light-dark transitions ~ in the second contrast class" and therefore does not correspond to the reference value indicating "100 light-dark transitions in contrast class 1". The deterioration of the possibility of recognition of defects by the device recognition then becomes noticeable by shifting the light-dark transitions from the contrast class 1 to the contrast class 2. By this version of the method according to the invention, statements can be made about the reliability of the recognition, in particular of small defects or defects in the form of transparent foreign bodies. With this version of the method according to the invention the characteristic signal _not only contain details of the. number of clear-dark transitions, but also details about the distribution of clear-dark transitions through the different kinds of contrast. The corresponding details are also contained in the reference value. By means of this, a particularly anticipated recognition of the reduction in the sensitivity of the recognition device is possible. A fouling of the lens system (s) of the recognition device does not lead first to a change in the amount of light-dark transitions, but certainly leads to a change in the distribution of the light-dark transitions between the different contrast classes. The adjustment of the reference values for an examination apparatus can be carried out manually, for example on a keyboard, or by means of the test objects same, by passing the test object through the examination apparatus and storing the confirmed characteristic signal as a reference value. It is also possible to determine the reproducibility percentage of the reference value, and thereby establish the maximum permissible deviation of the test signal above or below the reference value with which the second condition can still be considered satisfied, by additional steps of test object through the examination apparatus The method according to the invention is important in particular for the food industry, in that it is of paramount importance that the examination apparatus operate without errors and that a reduction in reliability of an examination apparatus is already recognized before the defective bottles, for example those bottles that still contain residues of caustic solutions pass without being recognized by the examination apparatus. It is equally important that in empty bottles there are no foreign bodies, such as dirt or cellophane films, that the internal pressure is not too high or too low, that the opening of the bottle does not have splinters and, in the case of cans , which are in a satisfactory state before being filled and sealed satisfactorily . The already described recognition process in which the number is determined and optionally the contrast of the light-dark transitions is particularly useful for recognizing foreign bodies in empty bottles and for recognizing splinters in the opening of the mouth. The absorbing foreign bodies, for example dirt, are established by the bright-field method, while transparent foreign bodies, for example films, are detected by the dark-field method (EP-A-0 387 930). The individual zones of an empty bottle and in particular those of the base of the bottle are examined separately. By means of the method according to the invention, the reliability of an examination apparatus can be verified separately by means of a single test bottle for each type of defect and each recognition zone, since a bottle can have a certain reference value for each type of defect and for each recognition zone. By this it is possible to greatly reduce the number of test bottles. A test bottle may have, for example, a defect in the mouth area and a defect in each individual recognition zone of the side wall, for example represented by a stop rod in the center of the bottle, and in each case a defect in the base area in the various recognition zones namely, the edge area, the transition zone, the center of the edge and the center of the base. The defects in the side wall and the areas of the base can be made of opaque insulating fabric and correspond to a defined reference value. Additionally, defects in the form of transparent foreign bodies, for example cellophane, can also be provided for the side wall and the base, in order to also verify the reliability of the examining apparatus with respect to the transparent foreign bodies. However, a separate test bottle is usually used to verify the reliability of an examination device that detects residues of caustic solutions in empty bottles. The caustic solutions are polar liquids, so they conduct high frequency electrical radiation better than air. Accordingly, residues of caustic solutions can be recognized by confirmation of the absorption of high frequency electromagnetic radiation. The characteristic signal is a measure of the lower absorption of electromagnetic radiation that is experienced by caustic solution residues. If the intensity of the electromagnetic radiation that an empty bottle allows to pass exceeds a threshold value, the first condition is no longer satisfied and the empty bottle in question is removed from the stages of Subsequent production as defective. The test bottle receives a certain amount of the caustic wash solution and the corresponding characteristic signal is entered into the test apparatus as a reference value. When the reliability of the residual caustic solution examining device is verified, the characteristic signal for the test bottle must correspond to the reference value with minor deviations. Simultaneously, the test of residual caustic solutions is an example of how the method according to the invention can also be used with analogous characteristic signals. A further example is the determination of liquid residues in the base of the bottle by measuring the attenuation of the resulting IR light. The residues of caustic solutions in an empty bottle are recognized both in the previous special test for residues of caustic solutions and in the general determination of liquid residues by IR attenuation. Therefore, certain characteristics of the objects can also be verified in two different ways. The method according to the invention can also be used to verify the reliability of examining apparatuses that work with different initial variables as a characteristic signal. Examples are systems for recognizing pixels (image elements) light or dark or systems to evaluate the distributions of brilliance, (histograms), which, considered by themselves or after the evaluation via arithmetic algorithms represent parameters for the quality of bottles or other objects. The attached figure shows a test bottle, which is recognized as defective both during the recognition of the lateral wall and during the recognition of the base and the mouth test, that is, it does not satisfy the first corresponding conditions. The bottle 10 is a normal water bottle of 0.7 1 made of glass. At the base the test bottle 10 has a radial pattern 12 which generates a very large number of light-dark transitions during the recognition of the base. Arranged axially in the center is a rod 14 that always generates the same number of light-dark transitions irrespective of the rotational position of the test bottle 10. Finally, the empty bottle 10 also has a spill 16 at the edge of the opening. The purpose of this splinter is to verify the reliability of the examining apparatus that inspects the mouth of the empty bottle by means of an image evaluation facility. The image evaluation installation comprises an outer annular recognition zone corresponding to the area of the mouth that curves towards outside, as well as an adjacent inner annular recognition zone corresponding to the horizontal area of the mouth. In the outer recognition zone the chip 16 leads to a decrease in the number of light-dark transitions, while leading to an increase in the number of light-dark transitions in the inner recognition zone. According to the method according to the invention, a test bottle of this kind is guided through the examining apparatus together with the flow of empty bottles. The test bottle must be able to be recognized as such so that the observance of the second condition can be verified in the examiner apparatus, namely the concordance with the predetermined or detected and stored reference value. For this purpose the test bottle has a mark, for example a metal ring or a ring made of triple-reflex metal tape, so that it can be recognized "as a test bottle by the examining apparatus by means of an appropriate recognition device. For example, a metal detector or a light barrier, but also the control device of the examination device can recognize an identification of the test bottle, by software (software), by the large number of defects that occur in this one. It would be very unlikely that a Such a large number of defects occurs in a normal empty bottle. If, therefore, the occurrence of all these defects in a single empty bottle is recognized by an order of magnitude that is typical for the test bottle, then it can be concluded that it is a test bottle and the observance of the second condition can be verified. .

The patent application entitled "Method for determining parameters, for example fill level, pressure or gas composition in closed containers" (internal reference: 31457 / Füllstandsk., Corresponding to DE 196 46 685.7) presented simultaneously describes a method to determine the level of filling of liquids in containers that are sealed by a lid, being that the lid is excited by a short magnetic pulse to carry out primary mechanical vibrations. The secondary vibrations excited by the primary mechanical vibrations of the lid of the container, which take place within the space between the lid and the liquid, are recorded by means of a microphone and analyzed, confirming the filling level by the established frequency of these vibrations high schools. The internal pressure of the container can also be deduced by an analysis of the primary mechanical vibrations. The method according to the invention can also be used on this occasion. On this occasion the signal characteristic is the frequency of secondary vibrations for the filling level. The first condition is that this frequency is above a certain value, for example 7 kHz; the lower frequencies mean that the air space in the neck of the bottle is too large and therefore the filling level is too low. The second condition, which is characteristic for the correct operation of the examiner apparatus, is that the characteristic signal for the test bottle corresponds to that of the predetermined reference value that was previously entered for the test bottle. Another possible way of verifying the reliability of an examination apparatus, and in particular the cleaning of the optical recognition system results when the wall contrast is examined during the recognition of side walls. The difference in contrast, ie the difference in brightness of the bottle edge that appears black due to the strong refraction of the light beams that penetrate it is determined in relation to the background illumination or in relation to the brightness of the center of the bottle The deviations from the average of this brightness difference allow a very anticipated statement about the deterioration of the recognition efficiency and consequently about the reliability of the examining device. The sensitivity of the examiner apparatus can be automatically tracked in a certain area by the previously described versions of the method according to the invention. In the case of relatively small deviations of the signal characteristic of the reference value, for example a deviation of 5%, the threshold values or limit values can be modified by a corresponding percentage, which are decisive when verifying the first condition. Only when the deviation is greater than, for example, 5% is that an alarm signal is given, and when an additional threshold value for the deviation, for example 20%, is exceeded, the examining apparatus stops because it no longer exists. there is a certainty of the recognition of specific defects. However, if the permitted deviation is established, the original detected and stored reference value is used in each case.

Claims (1)

1. CLAIMS Method for verifying the reliability of an examination device that inspects a large number of empty bottles of the same type, made of transparent material, with respect to the absence of defects and the absence of foreign bodies, in which an image of an empty bottle by means of a recognition device, a characteristic signal is generated for each empty bottle which is based on the distribution of the brightness (histogram), and the characteristic signal is compared with a threshold value, with a test bottle being leads to the examination apparatus after a number of empty bottles and that the characteristic signal of the test bottle is verified, characterized by the fact that when the characteristic signal of the test bottle is verified, it is verified whether the characteristic signal of the bottle test corresponds to that of a pre-established reference value. Method according to claim 1, characterized in that an alarm signal is given in the case of relatively small deviations of the signal characteristic of the test bottle of the reference value, and that the process of Inspection is interrupted in the case of major deviations. Method according to claim 1 or 2, characterized in that the reference signal is entered by storing the characteristic signal of a test bottle as a reference value. Method according to claim 3, characterized in that the allowed deviation of the characteristic signal of the test bottle from the reference value is determined by confirming the reproducibility of the characteristic signal by several steps of the test bottle through the examining apparatus, and then the allowed deviation is established so that the characteristic signals that were obtained during the various steps are within the allowed deviation. Method according to one of claims 1 to 4, characterized in that the characteristic signal based on the distribution of the brightness contains details of the number of light-dark transitions that occur. Method according to claim 5, being that the foreign bodies in the empty bottles they are recognized by the recognition apparatus, characterized in that the characteristic signal contains additional details about the distribution of the light-dark contrasts and that similarly the reference value contains details of the distribution of contrasts of the clear distributions. dark and, when the characteristic signal of the test bottle is verified, it is also verified if the details of the distribution of contrasts containing the characteristic signal agree with those of the reference signal. Method according to one of claims 1 to 6, characterized in that the reliability of one or more examination apparatuses is verified with a single test bottle when several characteristics of the empty bottles are verified. Test bottle for use in the method according to one of claims 1 to 7, characterized in that the test bottle is prepared in such a way that it always generates the same number of light-dark transitions irrespective of its position of rotation.