Device for optical inspection of parisons

A line for making containers of thermoplastic material, comprising:

• either one of i) a moulding machine configured to make parisons, or ii) a blow-moulding machine configured to receive the parisons and to blow-mould them in moulds to make the containers, and
• a device for optical inspection of parisons, the device for optical inspection of parisons including
• an illuminator including a light source configured to emit a beam of light directed towards a parison located at an inspection position;
• and
• a detector including a camera configured to capture an image of the parison located at the inspection position, wherein the parison, at the inspection position, is operatively interposed between the illuminator and the detector,
• the illuminator including an emission-polarizing filter configured to intercept the beam of light emitted by the light source and to generate a polarized light beam, and in that the detector includes a receiving polarizing filter configured to receive the polarized light beam, so that the parison, at the inspection position, is operatively interposed between the emission-polarizing filter and the receiving polarizing filter,
• the device for optical inspection of parisons further comprising
• a self-learning system;
• and
• a processing unit including
• a memory including reference data sets, and
• a processor programmed to process the image captured by the detector based on the reference data sets in order to derive a diagnostic indication regarding a defectiveness of the parison, so that the diagnostic indication distinguishes between a with-defect typology of objects and a defect-free typology of objects, the processing unit being trained, by the self-learning system, through samples of objects of the defect-free typology;
• wherein the processing unit is configured
• to process the image captured by the detector based on the reference data sets in order to derive from the image values of a plurality of image features,
• to process the values of the plurality of image features to derive the diagnostic indication regarding the defectiveness of the parison,
• to generate an image reconstructed from the values of the plurality of image features and based on the reference data sets, and
• to derive the diagnostic indication regarding the defectiveness of the parison as a function of comparing the image captured by the detector with the reconstructed image;
• and
• wherein the optical inspection device is operatively located either i) downstream of the moulding machine, or ii) upstream of the blow-moulding machine.

A method for optical inspection of parisons, comprising the following steps:

• emitting a beam of light directed towards a parison located at an inspection position, by means of an illuminator including a light source;
• capturing, with a detector including a camera, an image of the parison located at the inspection position, wherein the parison, at the inspection position, is operatively interposed between the illuminator and the detector;
• generating a polarized light beam by intercepting the beam of light emitted by the illuminator on an emission-polarizing filter interposed between the light source and the parison;
• and
• receiving the beam of polarized light on a receiving polarizing filter, interposed between the parison and the camera;
• wherein, at the inspection position, the parison is operatively interposed between the emission-polarizing filter and the receiving polarizing filter;
• the method for optical inspection of parisons further comprising a step of processing the image that includes sub-steps of
• processing the image captured by the detector based on the reference data sets, in order to derive from the image values of a plurality of image features;
• generating an image reconstructed from the values of the plurality of image features and based on the reference data sets;
• deriving a diagnostic indication regarding a defectiveness of the parison as a function of comparing the image captured by the detector with the reconstructed image;
• and
• training the processing unit through a self-learning system, so that the diagnostic indication distinguishes between a with-defect typology of objects and a defect-free typology of objects, the processing unit being trained, by the self-learning system, through samples of objects of the defect-free typology.

A method for processing an image of a parison, captured by a detector, the method comprising the following steps:

• processing the image captured by the detector based on the reference data sets, in order to derive from the image values of a plurality of image features;
• generating an image reconstructed from the values of the plurality of image features and based on the reference data sets;
• deriving a diagnostic indication regarding the defectiveness of the parison as a function of comparing the image captured by the camera with the reconstructed image;
• and
• self-learning, comprising the following sub-steps:
• capturing a plurality of images for a corresponding plurality of parisons;
• processing each image of the plurality of images based on the reference data sets, in order to derive from each image of the plurality of images corresponding values of a plurality of image features on the basis of a predetermined criterion;
• generating for each image of the plurality of images a corresponding reconstructed image using the corresponding values of the plurality of image features and based on the reference data sets;
• comparing each image of the plurality of images with the corresponding reconstructed image and deriving a corresponding similitude parameter representing a similarity between the image captured by the detector and the corresponding reconstructed image;
• updating the reference data sets and the plurality of image features as a function of the similitude parameter and of a predetermined threshold value;
• and
• training the processing unit through a self-learning system, so that the diagnostic indication distinguishes between a with-defect typology of objects and a defect-free typology of objects, the processing unit being trained, by the self-learning system, through samples of objects of the defect-free typology.

A method for processing an image of an object made of plastic material, the image being captured by a detector, the method comprising the following steps:

• processing the image captured by the detector based on reference data sets, to derive from the image values of a plurality of image features;
• generating an image reconstructed from the values of the plurality of image features and based on the reference data sets;
• deriving a diagnostic indication regarding a defectiveness of the object, as a function of comparing the image captured by the detector with the reconstructed image;
• and
• self-learning through the following sub-steps:
• capturing a plurality of images for a corresponding plurality of objects;
• processing each image of the plurality of images based on the reference data sets, to derive from each image of the plurality of images corresponding values of a plurality of image features on the basis of a predetermined criterion;
• generating, for each image of the plurality of images, a corresponding reconstructed image using the corresponding values of the plurality of image features and based on the reference data sets;
• comparing each image of the plurality of images with the corresponding reconstructed image and deriving a corresponding similitude parameter representing a similarity between the image captured by the detector and the corresponding reconstructed image;
• updating the reference data sets and the plurality of image features as a function of the similitude parameter and of a predetermined threshold value, and
• training the processing unit through a self-learning system, so that the diagnostic indication distinguishes between a with-defect typology of objects and a defect-free typology of objects, the processing unit being trained, by the self-learning system, through samples of objects of the defect-free typology.

A device for optical inspection of objects made pf plastic material, comprising:

• an illuminator including a light source configured to emit a beam of light directed towards a parison located at an inspection position;
• a detector including a camera configured to capture an image of the object located at the inspection position, wherein the object, at the inspection position, is operatively interposed between the illuminator and the detector,
• a processing unit including a memory, containing reference data, and a processor, programmed to process the image captured by the detector, based on the reference data, to derive from the captured image values of a plurality of image features, and to process the values of the plurality of image features, to derive a diagnostic information regarding a defectiveness of the object,
• the processing unit being configured for generating an image reconstructed, from the values of the plurality of image features and based on the reference data sets, and for deriving the diagnostic indication regarding the defectiveness of the object as a function of comparing the image captured by the detector with the reconstructed image;
• and
• a self-learning system configured
• to receive as input a plurality of images captured by the detector for a corresponding plurality of objects;
• to process each image of the plurality of images captured by the detector based on the reference data sets, in order to derive for each image corresponding values for the plurality of image features based on a predetermined criterion;
• to generate for each image of the plurality of images a corresponding reconstructed image, based on the reference data sets, using the corresponding derived values for the plurality of image features;
• to compare each image of the plurality of images captured by the detector with the corresponding reconstructed image in order to derive, for each image of the plurality of images, a corresponding similitude parameter representing a similarity between the image captured by the detector and the corresponding reconstructed image;
• and
• for each image of the plurality of images, to update the reference data sets as a function of the similitude parameter and of a predetermined threshold value for the similitude parameter,
• the self-learning system being programmed to train the processing unit so that the diagnostic indication distinguishes between a with-defect typology of objects and a defect-free typology of objects, the processing unit being trained, by the self-learning system, through samples of objects of the defect-free typology.