A method for inspecting absorbent articles is provided. The inspection is performed using an inspection algorithm generated with a convolutional neural network having convolutional neural network parameters. The convolutional neural network parameters are generated by a training algorithm. Based on the inspection, characteristics of the absorbent articles, such as defects, can be identified. Absorbent articles having identified characteristics can be rejected, or other actions can be taken.

A method for inspecting absorbent articles is provided. The inspection is performed using an inspection algorithm generated with a convolutional neural network having convolutional neural network parameters. The convolutional neural network parameters are generated by a training algorithm. Based on the inspection, characteristics of the absorbent articles, such as defects, can be identified. Absorbent articles having identified characteristics can be rejected, or other actions can be taken.

Absorbent articles comprising apertured regions and methods of manufacture and registration are disclosed. In an embodiment, a method of registering webs comprises moving first and second webs in a machine direction, capturing a first image comprising a portion of the second web, filtering the captured first image to determine one or more regions of the captured first image having a light transmittance value greater than the transmittance threshold value, modifying the captured first image with a dilation morphological operation, determining a feature of interest, determining a first difference value comprising a difference in location between the feature of interest and a reference feature of the captured first image or the modified captured first image, adjusting a speed of the second web in the machine direction, and coupling the first web and the second web together.

A method for inspecting absorbent articles is provided. The inspection is performed using an inspection algorithm generated with a convolutional neural network having convolutional neural network parameters. The convolutional neural network parameters are generated by a training algorithm. Based on the inspection, characteristics of the absorbent articles, such as defects, can be identified. Absorbent articles having identified characteristics can be rejected, or other actions can be taken.

A method for inspecting absorbent articles is provided. The inspection is performed using an inspection algorithm generated with a convolutional neural network having convolutional neural network parameters. The convolutional neural network parameters are generated by a training algorithm. Based on the inspection, characteristics of the absorbent articles, such as defects, can be identified. Absorbent articles having identified characteristics can be rejected, or other actions can be taken.

A method for inspecting absorbent articles is provided. The inspection is performed using an inspection algorithm generated with a convolutional neural network having convolutional neural network parameters. The convolutional neural network parameters are generated by a training algorithm. Based on the inspection, characteristics of the absorbent articles, such as defects, can be identified. Absorbent articles having identified characteristics can be rejected, or other actions can be taken.

A method for inspecting absorbent articles is provided. The inspection is performed using an inspection algorithm generated with a convolutional neural network having convolutional neural network parameters. The convolutional neural network parameters are generated by a training algorithm. Based on the inspection, characteristics of the absorbent articles, such as defects, can be identified. Absorbent articles having identified characteristics can be rejected, or other actions can be taken.

The present disclosure relates to systems and processes for inspecting and evaluating qualities of printed regions on substrates, wherein the systems and methods may be configured to eliminate subjective aspects relating to human involvement in performing visual checks when evaluating print quality. The systems may include one or more communication networks connecting one or more sensors with an analyzer. In operation, ink is applied to a substrate to create at least one printed region, and the sensors are configured to inspect the printed region. The sensors may be configured to communicate measurements and/or images to the analyzer. And the analyzer may then calculate one or more quality subscores based on respective measurements and/or images. In turn, a full print quality score may be calculated based on one or more of the quality subscores. The analyzer may then execute a control action based on the full print quality score.

A method is presented of forming protrusions in a portion of a substrate. The method may include providing a first device comprising an outer surface; providing a second device including a source of vibration energy; forming a nip between the source of vibration energy and the outer surface; conveying the substrate through the nip; forming the protrusions in the portion of the substrate in the nip using the source of vibration energy; and inspecting one or more characteristics of the substrate using a vision system.

A method for in-line analysis of a composite product, wherein a hyperspectral sensor is used to acquire images of samples of target materials that are part of the composite product, in order to perform an in-line optical inspection at process speed.