Method and related numerical control equipment for the automatic cutting of pliable sheets, for example for shoe, leather, clothing components and the like
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current automated cutting systems for pliable materials, such as shoe uppers or leather items, face challenges in precisely cutting portions with varying colors, iridescent effects, and complex shapes due to the difficulty in recognizing and aligning internal elements, leading to suboptimal cutting precision and potential assembly issues.
Innovation Solution
A method using a numerical control machine with an electronic control system that captures images of the material, filters them based on qualitative aspects like color and surface treatment, and adjusts the cutting path through a neural network to ensure precise alignment and cutting, allowing for efficient trimming without manual positioning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If automated cutting systems use peripheral references or markers on sheet surfaces for alignment, then cutting precision can be improved for standard materials, but the system cannot process new types of sheet materials with indistinct shapes caused by color shades and iridescent effects
Solution Approach 1:
The patent replaces traditional mechanical/optical marker-based alignment systems with an image processing and neural network-based system. The system captures images of the sheet material, processes them through neural networks to identify critical features and determine cutting paths, eliminating the need for physical markers or peripheral references on the material surface.
Solution Approach 2:
The patent changes the approach from relying on physical markers to using digital image parameters and neural network processing. By capturing images and analyzing color shades, iridescent effects, and surface features through software algorithms, the system adapts to various material types without requiring physical modifications to the sheets.
2Adaptability or versatility
If automated cutting systems process sheets with varying colors, iridescent effects, and complex shapes, then material versatility is improved, but the ability to recognize and align internal elements deteriorates
Solution Approach 1:
The patent replaces traditional optical alignment systems with neural network-based image processing. The neural network analyzes captured images to identify critical features and determine cutting paths, overcoming the limitations of conventional systems that cannot handle complex visual patterns like iridescent effects and varying colors.
Solution Approach 2:
The patent introduces an intermediary layer of image processing and neural network analysis between the camera and the cutting system. This intermediary processes the visual information, extracts meaningful features, and translates them into cutting path instructions, bridging the gap between complex material appearances and precise cutting requirements.
3Productivity
If automated cutting systems use traditional image processing methods, then processing speed is maintained, but cutting precision for complex materials with surface variations deteriorates
Solution Approach 1:
The patent replaces traditional image processing algorithms with neural network-based processing. The neural network automatically learns and identifies critical features in the captured images, providing both high processing speed and high precision for complex materials with surface variations like color shades and iridescent effects.
Solution Approach 2:
The neural network system is self-learning and self-adjusting, automatically adapting to different material types and surface characteristics without requiring manual programming or adjustment. This enables the system to maintain high processing speed while achieving high cutting precision across various material types.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The inventionrelates to amethod and to arelated numerical control machine for automatically cutting sheets of pliable material, for example for shoe, leather, clothing components and the like, not necessarily made of leather. The method of the invention is applied to a numerical control machine (7) provided with an electronic control system (20) having a memory (21) comprising at least one reference image (32) of the shape of the component to be cut, the machine being designed to receive at its inlet (5) material in the form of sheets (2) so as to define on said sheets (2) a cutting perimeter or trimming and to transfer cut portions (3) to its outlet (13). The method comprises at least the steps of: - storing in the memory (21) at least one second reference image (16) representing a qualitative aspect of the surface (24) of the sheet (2) or of the portion (3) to be cut; - defining by means of selection, via a user interface (30) of the electronic control system (20), at least one first sample portion (18) and at least one second sample portion (19) of said at least one second reference image (16), obtaining a criterion for comparative classification of said sample portions (18, 19) of the image; - acquiring at least one image (12) of the sheet (2) to be cut, by means of image detectors (10) associated with the machine (7); - classifying pixels of the acquired image (12) of the sheet (2) to be cut, assigning thereto a percentage value or weight based on said classification criterion, by means of an image processing unit of said electronic control system (20), obtaining a filtered image (28) based on this criterion; - further processing said filtered image (28) and the reference image of the shape of the component to be cut, so as to define the perimeter (29) of a cutting area of the portion (3) of the sheet (2).