Video tutorial: http://www.zmotion.com.cn/video/yemian/tvideos29.html
Today, Zheng Motion Technology will share with you "Quick Start Guide to VPLC Series Machine Vision Motion Control All-in-One Machine (Part 1)".
In recent years, with the advancement of science and technology, artificial intelligence, one of the three cutting-edge technologies of the 21st century, has developed rapidly. Visual technology, as a branch of artificial intelligence, has also developed accordingly, and various industries are gradually undergoing technological transformation. Some traditional industries that required manual inspection are beginning to adopt automated intelligent inspection methods. For example, cameras are used to replace human eyes in observing and inspecting objects; program algorithms are used to simulate human consciousness for judgment, analyze and reason about image information, and obtain image processing results.
Zheng Motion Technology will fully support the development of the machine vision automated inspection industry with its independently developed VPLC vision motion controller, providing machine vision + motion control solutions for professionals in various industries.
When a camera captures an image, the light emitted from the image passes through the lens and forms an image on the CCD. When the light reaches a certain pixel on the CCD, a corresponding charge is generated according to the intensity of the light. The magnitude of this charge is read as an electrical signal, and the intensity of the light on each pixel (i.e., the pixel density value, also called the grayscale value) can be obtained.
The working principle of machine vision is to use optical systems and image processing equipment to simulate human visual functions, extract information from the acquired target image and process it to obtain the required information about the object to be detected, analyze and judge it, and then inform the hardware equipment of the final result to guide the next action of the machine.
A complete machine vision system consists of multiple modules, generally including an optical system (light source, lens, camera), an image processing system, and an interactive interface.
For machine vision projects, image input is a prerequisite for detection, and image input is inseparable from industrial cameras. A camera is a tool that converts light signals into digital or analog signals. It is an important device for acquiring images and a core component of a machine vision system.
Cameras are widely used in various fields, especially for production monitoring, measurement tasks, and quality control. Industrial digital cameras are typically more robust and durable than conventional standard digital cameras. The industrial production environment dictates that industrial cameras must be able to withstand various complex and changing external influences, such as operating in harsh environments with high temperatures, high humidity, and dust.
1. Resolution
Resolution represents a lens's ability to record detail, measured in lines per millimeter (lp/mm). Higher resolution results in a sharper image. Display resolution, as commonly referred to, is the precision of the screen image, indicating the number of pixels that can be displayed. Higher resolution means a larger amount of image data; generally, cameras with high resolution don't have very high frame rates. Common camera pixel resolutions are as follows:
640 x 480 = 307200 (300,000)
1280 × 1024 = 1310720 (1.3 million)
1600 × 1200 = 1,920,000 (2 million)
2560 × 1920 = 4915200 (5 million)
3672 × 2754 = 10112688 (10 million)
2. Pixel Depth
Pixel depth refers to the number of bits used to represent each pixel. A common bit depth is 8 bits. Each pixel in a color image is represented by three components: R, G, and B. If each component uses 8 bits, then a pixel uses a total of 24 bits, meaning the pixel depth is 24. Each pixel can represent one of 2^24 colors. In this sense, pixel depth is often referred to as image depth. The more bits used to represent a pixel, the more colors it can represent, and the greater its pixel depth.
3. Maximum Frame Rate / Line Rate
The rate at which a camera acquires and transmits images is typically measured in frames per second (Frames/Sec.) for area scan cameras and in lines per second (Lines/Sec.) for line scan cameras.
4. Exposure mode and shutter speed
Line scan cameras typically use line-by-line exposure, allowing for selection of fixed line frequency and external trigger synchronization acquisition methods. Exposure time can be consistent with the line period or a fixed time can be set. Area scan cameras offer several common methods, including frame exposure, field exposure, and rolling line exposure. Digital cameras generally provide external trigger image acquisition functionality. Shutter speeds can typically reach 10 microseconds, and high-speed cameras can achieve even faster speeds.
5. Spectral Range
This refers to the sensitivity of the pixel sensor to different light waves. The typical response range is 350nm-1000nm. Some cameras have a filter added in front of the target to filter out infrared light. If the system needs to sense infrared light, the filter can be removed.
6. Interface Type
It has a Camera Link interface, an Ethernet interface, a 1394 interface, a USB interface, etc.
The VPLC516E from Zheng Motion Technology is a vision motion controller with an embedded open Linux system. It is the first all-in-one machine in China that integrates embedded motion control and machine vision.
ZDevelop is a debugging software independently developed by Zheng Motion Technology. It allows users to easily configure the controller, quickly develop applications, and perform real-time debugging of the program running on the motion controller. ZDevelop software can connect to the controller via serial port or network port. When downloading the program to the controller, users can choose between ROM power-off save download or RAM power-off non-save download. The program saved by power-off can be run directly offline.
It supports four programming methods: Zvision, ZBasic, ZPLC ladder diagram, and ZHMI configuration interface. Programs written using ZDevelop software can be downloaded to the positive motion controller or simulated on a PC platform.
ZVision Basic is a vision programming language independently developed by Zheng Motion Technology. Vision programs are developed in ZDevelop software using vision commands. After the project program is written, it can be directly downloaded into the VPLC controller for operation.
ZVision Basic language has a standard naming convention, is simple and easy to understand, and can complete the implementation of vision functions with short vision instructions. It currently supports a variety of vision operators such as vision positioning, vision measurement, data code recognition, blob block detection, and defect detection.
The following is a simple example of image processing. First, using the image acquisition interface in ZDevelop, a network camera is used to capture images in real time. After acquiring the images, simple thresholding is performed to mark the regions containing objects. Then, the number of parts in these regions is counted.
(1) First, create a new hmi interface in ZDevelop, create an image element window, and set the relevant latch channel number.
(2) Acquire images. If continuous image acquisition is required, an image acquisition loop should be established and used in the loop.
(3) Simple image processing. The acquired image is saved to the GrabImg variable. Further processing is then performed on the image, such as binarization, dilation, erosion, and opening/closing operations, to achieve a clear image quality. In this example, a monochrome camera is used. If a color camera is used, the original image needs to be converted to a single-channel grayscale image using ZV_RGBTOGRAY. The entire image region is then generated, and binary processing is performed on the entire image region to extract the black parts and store them in the region variable.
(4) Next, use the ZV_RECONNECT instruction to calculate and generate a list of connected regions. Use the ZV_REFILTER instruction to extract regions with suitable areas from the list, eliminate small noise points that cause interference, extract each region and draw it using the ZV_REGION instruction.
(5) The quantity of parts is obtained by using ZV_LISTCOUNT, which retrieves the quantity from the region list into the register TABLE array TABLE(0).
ZVOBJECT ShowImg,ColorGrabImg,Region,MASk,RegionConnecte,BinaryImage,Region_Id
DIM Count,ShowString(100),File_Name(100)
Count=0
while 1
if Count=6 then
Count=0
end if
File_Name=TOSTR(Count,1,0)+".bmp"
ZV_READIMAGE(ShowImg,File_Name,1) 'Reads a local image
ZV_LATCH(ShowImg,0) 'Displays the image
ZV_REGENFULLIMG(ShowImg, Mask) 'Full image region processing
ZV_RETHRESH(ShowImg, Mask, Region, 0, 128) 'Binarizes the image within the region specified by Mask, generating the binarized region.
ZV_REFILLUP(Region,Region) 'Fills holes and removes white parts from the part.'
ZV_RECONNECT(Region,RegionConnecte) 'Calculates the connected components of the region.
ZV_REFILTER(RegionConnecte,0,1500,15000,0) 'Filters the regions in the region list, keeping regions with an area between 1500 and 15000; regions with an area outside this range are filtered out.
ZV_LISTCOUNT(RegionConnecte, 0) 'Gets the number of connected regions in the list.
ZV_GRAYTORGB(ShowImg,ShowImg) 'Converts grayscale image to RGB image
DELAY (100)
PRINT "Number of parts: " tostr(TABLE(0),1,0) 'Print output result
for i=0 to TABLE(0)-1 'Iterate through and display the output images
ZV_IMGCOPY(ShowImg,ColorGrabImg)
ShowString = TOSTR(i+1,1,0)
ZV_LISTGET(RegionConnecte,Region_Id,i) 'Retrieves the element with index 0 in the list.
ZV_REGION(ColorGrabImg,Region_Id,0,zv_color(0,0,255)) 'Draws red for each region
ZV_TEXT(ColorGrabImg,"Count of parts:"ShowString,10,10,120,ZV_COLOR(255,255,0))
ZV_LATCH(ColorGrabImg,1)
ZV_CLEAR(ColorGrabImg)
DELAY (50)
next
DELAY (100)
Count = Count + 1
wend
This article is original content from Zheng Motion Technology. We welcome everyone to reprint it for mutual learning and to jointly improve China's intelligent manufacturing level. Copyright belongs to Zheng Motion Technology. Please indicate the source if you reprint this article.
