A device for inspecting and controlling the density of a moving web of cloth in a production line
Technical Field
This invention relates to a device for inspecting the density of cloth and, in particular, for inspecting the density of a web of moving cloth in a production line.
The invention also relates to a device for controlling the density of the web of moving cloth in a production line.
Background Art Cloth is used worldwide in the manufacture of garments such as teeshirts, trousers, coats, socks, and the like, as well as in other applications, such as the manufacture of curtain, carpet, and upholstery materials.
The manufacture of raw cloth has traditionally been carried out for many millennia by hand. However, large-scale manufacturing has now become a high precision automated process, with increasing speed, accuracy and quality being achieved. Different customers require cloth having different specifications, i.e. of colour, pattern and density. The density of the cloth is frequently specified in terms of the number of stitches, or courses er centimetre (CPC).
Given the demand for raw cloth, it is not unusual for a raw cloth manufacturer to produce many kilometres of clothier day. Therefore the
precise density of cloth produced is of great importance to the manufacturer. If the density is too high, the manufacturer will have wasted a substantial amount of material, whereas, if the density is too low, the customer will be dissatisfied. For these two reasons, it is necessary for the manufacturer not only to monitor, but also to control the density of the cloth as it is being produced.
At present, to adjust the density of cloth, manufacturers make widespread use of a type of machine known as a compactor, which uses a system of rollers and moving knives to either compact or stretch cloth as it passes through the compactor.
However, the compactor must be manually adjusted by an operator on a continual basis. In practice, the operator inspects the finished cloth at the end of the production line, using a magnification glass with graduations, in order to measure the density of the cloth. Depending on the result of the inspection, the operator will then adjust the compactor to either compact or stretch the cloth, and then repeat the inspection, and re-adjust the compactor, until the cloth is of a satisfactory density.
The above method has a number of disadvantages. It is tedious for the operator, and can result in eye and/or back strain. It is slow, taking up a substantial portion of thfe operator's time. There is a considerable time delay in achieving a satisfactory cloth density. The inspection is not very precise, as calculation of the cloth density will vary from one operator to another. Furthermore, as inspection and control cannot be carried out at a high frequency, the response time to a change in cloth density is slow.
It is an object of the present invention to overcome the disadvantages of manually inspecting the cloth and manually adjusting the compactor iteratively to achieve the required density.
Disclosure of Invention Thus, the invention provides a device for inspecting the density of a moving web of cloth in a production line, said device comprising a camera and a data analyser, the camera, in use capturing images of the cloth, and the images being immediately processed by the data analyser to determine the density of cloth in real time. The advantage of a device for inspecting the density of cloth in accordance with the invention is that an operator no longer has to perform the tedious inspection process as described hereinbefore. Furthermore, as the cloth density inspection, performed by the device, is consistent and automated, operator subjectivity is eliminated, therefore providing a more consistent and accurate determination of the density of the cloth.
Preferably, the camera operates at a shutter speed greater than 3,000 f ames er second.
Further, preferably, the camera operates at a shutter speed greater than 4,000 frames per second.
The advantage of a camera operating at a high shutter speed is that the images produced will be sharp enough to be processed by the data
analyser to accurately calculate the cloth density. The high shutter speed also enables the system to identify changes in the cloth density rapidly.
Preferably, t ie camera has a high magnification lens.
A high magnification lens enables the camera to identify detail, which is not visible to the naked eye, thus increasing the accuracy of the cloth inspection.
In one embodiment of the invention, the camera has an automatically adjusting aperture.
A camera with an automatically adjusting aperture will provide images of similar quality by compensating for any change in the ambient light or cloth colour. This enables the same image processing techniques to be used on each image.
Preferably, t ie data analyser applies a brightness compensation factor to each image. Applying a brightness compensation factor to the image in the data analyser coupled with the functioning of the automatic aperture in the camera results in processed images, which have similar characteristics notwithstanding differences in the actual brightness of each particular cloth sample. Further, preferably, the data analyser calculates the cloth density in terms of the number of stitches per cm of cloth length and/or width.
Thus, the cloth density can either be measured along the length of the cloth, across the width of the cloth, or over an area of the cloth surface.
Preferably, the data analyser identifies stitches in the cloth by digitising the images and applying a binary threshold thereto.
Alternatively, the data analyser identifies stitches in the cloth by applying a pattern-matching algorithm to the images.
Once the stitches have been identified, it is then a simple process to calculate the number of stitches per cm of cloth length and/or width. Preferably, the data analyser can perform an error check to eliminate incorrectly identified stitches.
The advantage of eliminating incorrectly identified stitches is that the accuracy of* the calculated cloth density is increased.
Further, preferably, the data analyser eliminates from its calculations any set of identified stitches equal to or less than a preset minimum number of stitches.
Suitably, the preset minimum number of stitches is in the range 2 to 5.
Preferably, the preset minimum of stitches is 3.
Thus, if the device finds five stitches side by side, the data analyser regards this set of stitches as forming a valid basis for calculating the average distance between stitches. If, however, the device only detects three stitches in the set, this set is rejected as there is a higher chance that the patterns identified are random and therefore not a valid set of stitches.
In a further embodiment, the cloth is illuminated by a high intensity light source located at a position adjacent the cloth.
Illuminating the cloth with a high intensity light source has the advantage of improving the quality of the images captured by the camera.
Preferably, the light for the high intensity light source is generated at a position adjacent to the cloth.
The advantage of this feature is that, when there is space available at the position adjacent the cloth, it is simple and cheap to design a device wherein the light for the high intensity light source is generated at the high intensity light source.
Alternatively, the light for the high intensity light source is generated at a position remote from the cloth and conveyed to the position adjacent the cloth via a fibre optic conductor.
This is advantageous as a bulky device, such as a laser, can be used to generate light for the high intensity light source. As the bulky light generator is remote from the high intensity light source the light
source need not itself be bulky. This is useful when there is limited space available to locate the high intensity light source at a position adjacent to the cloth.
Preferably, the high intensity light is generated by a low voltage source.
A low voltage source is advantageous as, in a production line setting, if an accident happens, a low voltage source is less dangerous than a high voltage source.
In a further embodiment, the high intensity source is a plurality of light emitting diodes (LEDS).
The use of a plurality of LEDS as the high intensity light source is advantageous as LEDS are robust and require little maintenance.
Preferably, the light from the LEDS is pulsed for short periods of time, with each pulse of light being synchronised with the opening of the camera shutter.
The pulsing of the light from the LEDS results in further intensification of the light leading to improved image capture.
Preferably, the high intensity light source is directed towards the cloth surface at an angle to the perpendicular of the surface of the cloth in the range 30° to 60°.
More preferably, the high intensity light source is directed towards the cloth surface at angle to the perpendicular of the surface of the cloth in the range 40° to 50°.
A light source, directed at an angle off the perpendicular from the surface of the cloth, has been found to result in a change in the quality of the images, which has the effect of improving the identification of stitches. An angle of 45° has been found to be very effective in improving identification of stitches in herringbone patterns.
Suitably, the intensity of light emitted from the high intensity light source is automatically adjustable in response to the brightness of the cloth, such that, in use, the brighter the cloth the lower the light intensity and vice versa.
Matching the intensity of the light to the brightness of the cloth increases the accuracy of the device and is particularly advantageous where the cloth being measured is highly patterned in contrasting colours.
In a further embodiment, the moving web of cloth passes a roller, located adjacent the camera, the roller rotating at an adjustable speed such that the web of cloth passing over the camera is maintained at a uniform tension over time.
When a section of the cloth is stretched, the surface area of the cloth will increase, but the number of stitches will stay the same. Therefore, the more the section of cloth is stretched, the lower the
density will be. Variations in tension can thus lead to variations in the cloth density calculations, and impinge on the accuracy of the calculations. Therefore, it is important to maintain the moving web of cloth passing over the camera at a uniform tension over time. Preferably, a plurality of sensors, located above the web of cloth, measure the distance from the sensors to the web of cloth, the sensors providing instructions for automatically adjusting the roller on a continual basis, such that the web of cloth is maintained at a uniform tension over time. Measuring the distance between the moving web of cloth is an effective method of determining the tension of the web of cloth, as with decreases in tension, the cloth will slacken and drop away from the sensors, and with increases in tension, the cloth will tighten and rise towards the sensors. In a further embodiment, a sensor is mounted on a flywheel, the flywheel being in communication with the web of cloth, and the sensor being in communication with the data analyser, such that the rate of cloth production, and the length of cloth produced can be recorded as a function of time. Preferably, the cloth density, as a function of the length of cloth produced, is recorded on a remote hard drive in a local area network (LAN).
Alternatively, the cloth density, as a function of the length of cloth produced, is recorded on a hard drive in the data analyser.
The output from the sensor allows the operator to view the actual cloth density as a function of either the rate of cloth production and/or the length of cloth produced at any time or date. This is advantageous as it helps to monitor the device, and is also useful for quality control.
In a further embodiment, the camera and/or the data analyser are housed within a cabinet.
The cabinet protects the system from accidental impacts. Preferably, the cabinet is portable and is in communication with a portable flat screen monitor on which the images from the camera and the output f om the data analyser can be viewed.
The advantage of a portable cabinet is that it can be moved to any location in the cloth-manufacturing shop floor for random or systematic inspection of a particular web of cloth.
Preferably, a value specifying the required cloth density is input to the data analyser by a user and is stored therein.
The advantage here is that the user can easily compare the required cloth density with the actual cloth density, as calculated by the data analyser.
Preferably, the density of the cloth calculated by the data analyser is compared with the required cloth density on a continual basis.
This has the advantage of automating the comparison between the required cloth density and the actual cloth density, removing the possibility of human error.
In a further embodiment, the data analyser is in communication with a cloth compactor, the data analyser providing instructions for automatically adjusting the cloth compactor on a continual basis such that the density of the cloth produced converges towards the required cloth density.
As the device is in continuous communication with the compactor, a change in the cloth density will immediately trigger the data analyser to send instructions to the compactor to counteract the cloth density change. The risk of human error in adjusting the compactor is removed, and there is also no longer any risk of varying cloth density dμe to different operators.
Mode for carrying out the Invention
The invention will be forther illustrated by the following description of an embodiment thereof, given by way of example only with reference to the accompanying figure which is a schematic representation of a device for inspecting the density of a web of cloth in a production line according to the invention.
Referring to the figure there is illustrated, generally at 10, a device for inspecting the density of a moving web of cloth 11 in a production line comprising, a camera 12 and a data analyser 13, the
camera 12, in use capturing images of the cloth 11, the images being immediately processed by the data analyser 13 to determine the density of cloth 11 in real time.
The camera 12 is high speed, operating at 4,000 frames per second allowing the camera 12 to acquire sharp images of the cloth 11 at foil production speed, without any significant blurring of the images. The camera 12 has a high magnification lens 14, enabling the camera to acquire images of minute detail of the cloth 11. The camera 12 also includes an automatically adjusting aperture (not shown). This means that, when production changes f om a cloth of one brightness level to another, or there are changes in the ambient lighting, the camera 12 will automatically adjust its aperture, thereby producing images of similar brightness and contrast.
The images are provided to the data analyser 13, where stitches are identified, by digitising the image and applying a binary threshold thereto. Alternatively, the stitches can be identified by applying a pattern-matching algorithm thereto. The data analyser 13 can also eliminate incorrectly identified stitches.
The cloth 11 is illuminated by a high intensity light source 15, directed towards the surface of cloth 11 at an angle of 45° to the perpendicular of the surface of cloth 11.
Two sensors 16, 17 measure the distance of the cloth 11 from the sensors 16, 17 after it has passed over the device 10. The sensors 16, 17 are in communication with the data analyser 13, which issues commands to either speed up or slow down a roller 18, in such a way that the vertical position of the section of cloth 11, passing under the sensors 16, 17 is constant, and thereby insures the section of cloth 11 passing over the device 10 is maintained at a constant tension over time.
Another sensor (not shown) is mounted on a flywheel (not shown), which is in communication with the web of cloth 11. The sensor is in communication with the data analyser 13 such that the length of cloth 11 produced can be recorded either on a remote hard drive in a LAN or internally in the data analyser 13.
The camera 12, lens 14 and light source 15 are housed within a cabinet 19, which in turn is mounted on a pillar 20. In use, the web of cloth 11 travels over the top of the cabinet 19, wherein the camera 12 records images of the cloth 11 through a window 21 in the cabinet 19. The window 21 is made of glass and prevents dirt falling on the lens 14.
Having identified the stitches, the data analyser 13 calculates the density of cloth 11. The density of cloth 11 is then compared with the required cloth density, which is input to the data analyser 13 by a user and is stored therein.
The cloth 11 also travels through a compactor 22, which can be adjusted to either increase or decrease the density of the cloth 11. The results from the comparison of the density measured by the device 10 and the density input by a user can then used to calculate the adjustment to be manually made to the compactor 22, to either compact or stretch the cloth 11, such that the density of the cloth 11 produced converges towards the required cloth density.
Alternatively, the cloth compactor 22 can be in direct communication with the data analyser 13, the data analyser 13 providing instructions to the compactor 30 to either compact or stretch the cloth 11. The instructions are based on the comparison between the cloth density calculated by the data analyser 13 and the required cloth density, such that the density cloth 11 produced at any given time converges towards the required cloth density.