Understanding And Applying Machine Vision Part 11 pot

Similarly, a machine vision system might have to somehow include a weighting factor e.g., a complementary color detector that will increase sensitivity to a specific factor - color in th

Long Line Life

Installations of the machine vision system should be associated with a new model or one that has been introduced recently This should guarantee the payback from the system Ideally, the system should be incorporated with new tooling

10—

Operation Champion

The plant selected for the installation should have someone, preferably in manufacturing engineering, who wants to see

it work and will ensure it does

Having established the objectives along the lines of those depicted in Chapter 4 for the project, the next step is to develop a set of detailed specifications for the application Specifications are technical data that describe all the

necessary functional characteristics that the system must have to perform the required job They should not state how the vendor is to do the job Specifications must state the required productivity and capability of the system as well as all significant operational requirements They should also explain the requirements of the related fixtures, material handling, and so on Similarly, requirements for machine control and line interfaces should be detailed as well as plant requirements and limitations

The importance of specifications cannot be overstressed for the time and effort put into proper specifications will be more than repaid in terms of reduced start-up time and a reduction in maintenance and quality problems

By analyzing the present methods, observing the present operation, and getting input from all those involved, a

detailed description of the operation and a

Page 316review of all anticipated variables can be developed For example, it may appear that an operator is simply performing

a sorting function, separating containers by their size or shape On closer scrutiny, one will also observe that once in a while the operator throws a container into a scrap bin While probably not classified as an inspector, by virtue of innate intelligence, the operator knows enough to separate incomplete or misshapened containers

By virtue of increased sensitivity to exceptions, an operator can become more sensitive to specific conditions For example, because training has emphasized the importance of separating all green containers, the person has a

heightened awareness when such objects pass Similarly, a machine vision system might have to somehow include a weighting factor (e.g., a complementary color detector) that will increase sensitivity to a specific factor - color in this example

The converse is also true An operator may be trained to be tolerant of color shade variations: For example, all yellows, regardless if pale gold or virtually orange, might be acceptable since it is the basic color itself that provides the

distinction In this case, therefore, the machine vision system must be equally tolerant of these normal variations while also maintaining sensitivity to the fundamental defects the system is supposed to capture

Developing the specifications includes a quantification of existing procedures in terms of productivity and quality: number of shifts, scrap, warranty repairs, machine downtime, and so on

When examining the operations, the following should be considered:

How are goods routed for the vision inspection task presently?

Does the operation require automating loading/unloading?

Is the operation inventoried?

Are products stored in bins? Magazines? And so on?

What are the actual inspection functions that must be performed by the system?

Is it gauged by "eyeball" or with instruments (micrometer)?

Is cosmetic inspection done by detail or is it performed by a cursory look for more gross appearance differences?

Does the inspection first require identification? If so, is it by shape? By reading characters? And so on?

Is the inspection itself one of verifying shape conformance? Again, with or without instruments?

Is the operation one of just verifying that an assembly is complete before another operation is performed or that objects are oriented properly?

If, in fact, it is a complete assembly operation that is performed, does the operator locate parts and guide them into place?

Is this operation a combination of several of the preceding tasks: location guidance, cosmetic examination, and

gauging, for example? Is the present task 100% inspection or sample inspection?

Page 317Does the operator perform tasks other than vision (e.g., assembly and machine loading)?

Is there contrast in what must be observed, that is, can you visually see the condition without picking up the part to manipulate it in the light? How small is the smallest detail you want to see? How big is the object (field of view)? As

in photography one can see both large and small objects with television cameras, but the detail that can be detected reliably is proportional to the field of view Where necessary, machine vision systems can employ more than one camera so detail versus field of view need not be a factor that would preclude considering machine vision

Are there normal variations in the appearance of the object that are ignored? A vision system will somehow have to normalize those conditions Are there variations in the appearance of the background? Is the part repeatably and

consistently located? If not, image capture will dictate a requirement for an even larger field of view, reducing the detectable detail It will also require location analysis to reconcile the image captured with respect to position

Are parts overlapping? Touching? Jumbled? Can they be presented registered and not touching?

Are parts moving or indexed where they can be held in place in front of the scanner? If in motion, at what conveyor speed? How well regulated is the speed?

Other considerations are as follows:

Does the operator perform three-dimensional analyses? Is the decision based on color interpretation?

How much time is there to make a decision?

What of the performance of any system substituted for a person? What percentage of the reject objects will one tolerate

to pass as good? What false reject rate (number of good units that are rejected) will be allowed? No system is perfect!How much start-up time will be allowed? If the system is not dedicated forever to a specific task, how much time will

be permitted to get the system ready between product changeovers? How much floor space is available? Overhead space that may be needed to mount cameras and/or lighting arrangements? Will much equipment rearrangement be required? Are power, air, and so on available? How much downtime can be tolerated for other routine maintenance?

In what kind of environment will the equipment operate? In the presence of dirt and dust? Grease and lubricants? Water? Shock and vibration? Temperature and humidity? Electrical noise?

To support the specification of a system can the following be made available to the prospective vendors for bidding: job descriptions, present specifications, drawings, samples, photographs and/or videotapes of the facility and

inspection area? What kind of personnel will be available to operate and service the system following installation?

Page 318

If a system is in place today to perform the function, answer the following:

(a) What technology is used?

(b) What are the system's capabilities and limitations?

(c) What actual performance is being experienced?

(d) What problems are being experienced that hamper productivity and effectiveness?

(e) What is the utilization factor?

(f) If starting today, what would be done differently?

(g) What problems were experienced and how were they overcome?

(h) What impact did the project have on the user organization? Was it measurable? If not, why not?

(i) Is the impact more or less than target levels? Why?

(j) Does upper management perceive improvements attributable to the project?

(k) What is currently perceived as the system's greatest contribution or benefit? Is it the intended one? If not, why not?(l) Was the budget maintained? The time schedule? If not, why not?

The following can be used as a guide in the "system" analysis:

2 In-house skills required and available

a Ability to do material handling

b Ability to do installation

c Personnel available to operate

d Personnel available to service

(b) Review any drawings: equipment layout drawings (are there critical dimensions?), those that affect functional capabilities, and part drawings What are the tolerances?

(c) Job descriptions - do they agree with observations?

(d) Part specifications

(e) Samples - do they experience corrosion or discoloration or other change with time?

(f) Photos of the prospective installation site as well as parts

The following checklist can serve as a guideline in systematically examining the requirement as observations are made:

1 Scene complexity

(a) Number of stable states

(b) Number of parts in view at one time

(1) Touching(2) Evenly spaced(3) Registered(4) Overlapping(5) In a bin(c) Number of features and description

(i) Temperature

(j) Other

(k) Positional variations

(1) Registration (x, y, z, theta)

(2) Object distance(1) Part sensitivity to heat

(e) Indexing (triggering possibilities) still time

(f) Continuous (triggering possibilities)

Page 320

3 Line speed

(a) Regulated speed

(b) Synchronous operations

4 Perspectives (camera, subject)

(a) Multiple perspectives

(b) Variable perspectives

(c) Vision access

5 Background contrast good

(a) Backlighting possible

(b) Gray scale, front light

(c) Structured lighting possible

(d) Color interpretation required

(e) Three-dimensional and/or depth contrast

(f) IR and/or temperature contrast

(d) Height variations (depth of field)

(e) Other

7 For character-reading applications

(a) Reading or verification

(b) Inked, painted; molded, cast; stamped; raised; laser-etched; dot matrix; acid-etched; engraved; recessed; other

(c) Describe single-font style and type and multiple-font style and types for each font style:

(1) Stroke width(2) Aspect ratio(3) Character height(4) Character width(5) Character depth(6) Center-to-center spacing(7) Space between characters(8) Characters per string and per line(9) Numbers of strings and lines(d) Read rate (characters per second)

(e) Character positioning, repeatability

(1) Vertically(2) Horizontally(3) Skew

(f) If "no read," what to do?

(g) Percentage of misreads allowed per character string

Page 321

8 System performance

(a) Percentage of subjects flawed

(b) False rejects allowable

(c) Escape rate allowable (how many bad ones can be allowed to pass?)

(d) Warm-up time allowable

(e) Setup time permissible per batch (changeover time)

(f) Skill level of person performing changeover

(g) Training time allowed

(1) For new subjects(2) Modified subjects(3) Frequency of new and modified designs(h) System reliability

(i) Response if triggered with no part in view

9 Physical and interface requirements

(a) Floor space

(b) Camera location relative to the subject, feet per inches

10 Electrical

(a) Power available, power preferred

(b) Line conditioned and/or regulated

(c) EMI/RFI in environment

(d) Computers and program controllers

11 Interfaces

(a) Interconnected equipment

(b) Communications protocols required

(c) Trigger signals, time delays

(d) Reject signal delays

(e) Data log; specify details required (CRT display, hardcopy/printer, indicator lights)

12 Operator interface

(a) Preferred CRT/keyboard, menu, and icons

(b) Report generation, describe

(c) On-line programming skills available

(d) Setup and calibration

13 System reliability and availability

Page 322(a) Number of hours used per week (up time)

(b) Number of hours available for maintenance per week

(c) Maximum hours to wait for repair

(d) Redundancy, back-up required

(a) Layout drawings available?

(b) Material handling to be whose responsibility?

(c) Special enclosures to be specified

(d) Personnel available

(1) To operate(2) To service(e) Material handling

(f) Training: operator, service

(g) Service spares

(h) Future modifications and enhancements desired

Specification Review from Machine Vision Perspective

Having distilled the preceding and written a specification, give some thought to the application from the machine vision perspective and review and revise the specification accordingly The following can serve as a guideline

13.2.1.1—

Lighting

Is special lighting needed to provide even illumination? Will reflections be a problem? Are strobes required? Can lighting exaggerate contrast of attribute to be observed?

Gauging Exaggerate edges, structured edges, backlighting, directional light and/or shadow.

Flaws Dark- and light-field illumination, UV fluorescence.

Recognition Back illumination, structured light, directional light.

Identification Light trapped, light scattered.

Verification Structure, specular, directional.

Guidance/location analysis Backlighting, structured light.

13.2.1.3—

Sensors and Cameras

The significance of delectability versus resolution should be recognized, that is, the ability to observe a single object versus the ability to separate objects The relationship of pixel arrangement to minimum detail observable should be understood

Gauging may require use of line scanners to obtain required accuracy It may be necessary to use a multiple-camera arrangement to make differential measurements, and stereo techniques may be needed to compensate for magnification changes due to part motion in the field of view

Flaw detection is required for field-of-view analysis The smallest reliably detectable flaw will typically be on the order of 0.3% of the field of view, not the object size

For reliable verification or recognition the minimum size of the attribute upon which the decision should be based is 1% of the field of view assuming relatively high contrast associated with that attribute

13.2.1.4—

Preprocessing and Processing

Preprocessing and enhancement needs can be eliminated or reduced by the following conditions: registration, contrast (real or artificial), and windowing unwanted areas

The application dictates the amount of preprocessing required: noise removal, smoothing, thresholding, segmentation, and so on If segmentation is required, it is for the gray level, edges, color, texture, or changes between images For example, an analysis outlined in a 1984 Carnegie-Mellon University dissertation, entitled ''The Potential Societal Benefits from Developing Flexible Assembly Technologies," by Jeffrey L Funk, for location analysis is as follows (Bi, binary; ED, edge; GFA, global feature analysis; LFA, local feature analysis):

A Rigid parts (same stable state), repeatable position

Again dictated by application, image analysis refers to the specific features that contain the data upon which decision is based Is the number of features constant or do they vary by samples, batches, and so on? What will the program do if

a feature is missing or an unwanted feature is present? Can statistical decision-processing techniques work or does it require syntactic analysis?

13.2.1.6—

Mounting and Interface

Where will camera(s) and lighting be mounted and how? Do custom mounts need to be designed and built? What will produce the electrical trigger that initiates the vision system to capture and process the images? How will reject

products be removed from the line? When will they be removed, immediately or downstream?

Does the system need to communicate with printers? Computers? Controllers? Other equipment?

Access Controls and Visual Displays

Which plant personnel should have access to the system during setup and operation? Is there need for password and security system? Key lockout?

Should the system display inspection results for a run on demand? Should the display be a graph with icon-like

indicators of trends? Should the system monitor trends from nominal conditions to indicate shift in process that could lead to rejects, an indication that corrective action is needed before rejects are produced Should this be a bell?

13.2.2—

Writing Final Specification

Having evaluated the process from the different perspectives, one can finalize the specification Above all, the

specification should not tell the vendor how to do the job; it should not specify the design or image processing

techniques, and so on Rather it should specify the application The specification should include the following

1—

Scope

This section should include an overview of the existing process and the overall system requirements as well as what the vision system will be required to do Material handling and interface requirements to existing machines and control systems as well as future anticipated interfaces should be explained

Page 325

2—

Part Description and Specification

This should include the distillation of all the considerations reviewed in the preceding about the process, accept-reject criteria, escape rate-false error tolerances, environmental considerations, utilities, data storage and transmission,

operator interfaces, cycle times, changeover times, and so on

3—

Acceptance