A print data editing device includes a controller configured to performs acquiring print data, and converting the print data. The converting includes determining, as a conversion sub-dot, one or more of sub-dots included in a sub-line for each of all sub-lines constituting one line within printing areas in an image represented by the print data, and editing the print data such that the print data indicates OFF for the one or more sub-dots determined as the conversion sub-dot. Each sub-line is constituted by the sub-dots. The sub-dot is a printing unit obtained by dividing a dot into M parts in a sub-scanning direction, where M is an integer of two or greater. The sub-dot for which the print data indicates ON is a printing area. The sub-dot for which the print data indicates OFF is a non-printing area. An area outside a printing region is also the non-printing area.

The present invention relates to an apparatus (100) for printing an article, the apparatus (100) comprising: a data providing device (10) adapted to provide article-specific printing data for the article from product data of the article, and to provide machine-specific printing data for the article in dependence on a printing machine to be used for the printing process; and a control device (20) adapted to merge the article-specific printing data and the machine-specific printing data, and to control the printing machine for printing the article based thereon.

The technology relates to a memory insertion machine for inserting memory modules into memory sockets on a circuit board. The memory insertion machine may include one or more insertion rods moveably mounted to one or more vertical guides, one or more profilometers, and an insertion controller. The insertion controller may be configured to apply an insertion force to a memory module in a memory socket, by controlling the movement of the one or more insertion rods on the one or more vertical guides. The insertion controller may be further configured to determine, based on information received from the one or more profilometers, a measured distance between a top of the memory module and a top of the memory socket.

A print data editing device includes a controller configured to performs acquiring print data, and converting the print data. The converting includes determining, as a conversion sub-dot, one or more of sub-dots included in a sub-line for each of all sub-lines constituting one line within printing areas in an image represented by the print data, and editing the print data such that the print data indicates OFF for the one or more sub-dots determined as the conversion sub-dot. Each sub-line is constituted by the sub-dots. The sub-dot is a printing unit obtained by dividing a dot into M parts in a sub-scanning direction, where M is an integer of two or greater. The sub-dot for which the print data indicates ON is a printing area. The sub-dot for which the print data indicates OFF is a non-printing area. An area outside a printing region is also the non-printing area.

A method and system to generate training data for a deep learning model in memory instead of loading pre-generated data from disk storage. A corpus may be stored as lines of text. The lines of text can be manipulated in the memory of a central processing unit (CPU) of a computing system, using asynchronous multi-processing, in parallel with a training process being conducted on the system's graphics processing unit (GPU). With such an approach, for a given line of text, it is possible to take advantage of different fonts and different types of image augmentation without having to put the images in disk storage for subsequent retrieval. Consequently, the same line of text can be used to generate different training images for use in different epochs, providing more variability in training data (no training sample is trained on more than once). A single training corpus may yield many different training data sets. In one aspect, the model being trained is a deep learning model, which may be one of several different types of neural networks. The training enables the deep learning model to perform OCR on line images.

A method and system to generate training data for a deep learning model in memory instead of loading pre-generated data from disk storage. A corpus may be stored as lines of text. The lines of text can be manipulated in the memory of a central processing unit (CPU) of a computing system, using asynchronous multi-processing, in parallel with a training process being conducted on the system's graphics processing unit (GPU). With such an approach, for a given line of text, it is possible to take advantage of different fonts and different types of image augmentation without having to put the images in disk storage for subsequent retrieval. Consequently, the same line of text can be used to generate different training images for use in different epochs, providing more variability in training data (no training sample is trained on more than once). A single training corpus may yield many different training data sets. In one aspect, the model being trained is a deep learning model, which may be one of several different types of neural networks. The training enables the deep learning model to perform OCR on line images.

Example methods and systems are described in which a color space vector is transformed into a Neugebauer primary area coverage (NPac) vector, to be used for printing. In some examples, the color space vector is transformed into a NPac vector on the basis of criteria associated to amounts or probabilities of Neugebauer primaries (NPs).

A data generating apparatus includes a controller to perform a first process to generate and store into a work area n pieces of first-layer object data representing n objects having earlier orders than a combined object in a combining sequence, a second process to generate and store into the work area M pieces of second-layer object data representing M second-layer objects of the combined object, a third process to generate first bitmap data representing an image obtained by combining the n first-layer objects, using the n pieces of first-layer object data, and a fourth process to generate second bitmap data representing the combined object. When a free space is less than a reference value during the second process, the controller interrupts the second process, performs the third process or a part of the fourth process, releases a storage area for object data, and thereafter resumes and completes the second process.

The technology relates to a memory insertion machine for inserting memory modules into memory sockets on a circuit board. The memory insertion machine may include one or more insertion rods moveably mounted to one or more vertical guides, one or more profilometers, and an insertion controller. The insertion controller may be configured to apply an insertion force to a memory module in a memory socket, by controlling the movement of the one or more insertion rods on the one or more vertical guides. The insertion controller may be further configured to determine, based on information received from the one or more profilometers, a measured distance between a top of the memory module and a top of the memory socket.

A data generating apparatus includes a controller to perform a first process to generate and store into a work area n pieces of first-layer object data representing n objects having earlier orders than a combined object in a combining sequence, a second process to generate and store into the work area M pieces of second-layer object data representing M second-layer objects of the combined object, a third process to generate first bitmap data representing an image obtained by combining the n first-layer objects, using the n pieces of first-layer object data, and a fourth process to generate second bitmap data representing the combined object. When a free space is less than a reference value during the second process, the controller interrupts the second process, performs the third process or a part of the fourth process, releases a storage area for object data, and thereafter resumes and completes the second process.