Zia Vision | Online Help - Zoho CRM

Zia Vision — Intelligent Image Validation

You can use Zia Vision to automatically validate the images that are added to your CRM's records. Using this feature, you can ensure that these images are accurate, consistent, and of the requisite quality. This has multiple benefits, ranging from adhering to guidelines to providing a positive experience for all users of your CRM.

Scenarios
  1. A real estate company wants to verify that images of luxury villas show a swimming pool. They can use Zia Vision to meet this requirement.
  2. A PC company wants to standardize its product images and ensure that they always contain a keyboard, a monitor, and a mouse. Zia vision can help in this situation.
  3. A home theater company has recently stopped manufacturing speakers as part of cost-cutting measures. They don't want images of their home theater systems to have speakers in them. They can use Zia vision to ensure this.
  4. An electronics company deals with the sale of household appliances, such as televisions, washing machines, mixers, and blenders etc. When they use the CRM, they need to upload images of products from multiple brands into their modules. The company uses Zia Vision to ensure that these product images are accurate and consistent, as these images could influence a sales person's conversation with customers and, subsequently, a customer's decision to buy the product.

Availability

Users need to have the Image Validation permission enabled for their profile. This permission is present under Setup permissions > Zia for each profile.
Under the Image Validation permission, you have the following options:

  1. Manage Configuration: Users can create, edit, view, enable, disable, or delete a rule. This has to be enabled manually.
  2. Manage Action: If images are declared as invalid by Zia, they are sent for manual approval. Users with this permission can approve or reject those images. When you enable the Image Validation permission, this option gets enabled by default.
If both these options are enabled for a profile, its users can both manage the rules and review the images invalidated by those rules.


How Zia Image Validation Works

To validate images automatically, Zia needs to have an idea of what "valid" or "invalid" means. Since Zia is an AI engine, she needs to be trained with several samples of desired (or undesired) images. For example, we could validate property listing images by training Zia with images like these:

Once trained, Zia will validate incoming images based on what she's learned.

Invalid images can be reviewed manually to ensure that valid images are not mistakenly removed. To improve her performance, you can modify the training data and retrain Zia.

Validation types

Validation of an image happens through the creation of an image validation rule. For each rule, Zia can match, detect, or match and detect entities in incoming images for one image field.

Entities that take up over 80% of the image need to be matched, while those under that threshold need to be detected. For example, when validating images of living rooms, the living room needs to be matched while a television or a table within it needs to be detected.

We use the term labels to denote entities that need to be matched. For example, the living room in the image below. The term objects is used to denote entities that need to be detected. In the image below, the television, vase, and table are objects.

You can pick one of the validation types based on your needs:
  1. Match only: Ideal when you want the validation to be done based on a complete match. For example, when you want to ensure that only images of the exterior of the house are used for property listings.
  2. Detect only: Ideal when you want the validation to be done based on a specific object detected in the image. For example, when you want to make sure that a swimming pool is present in all the images of luxury villas.
  3. Match and detect: Ideal when you want to validate based on both the complete image and the objects detected in it. For example, when you want to make sure that images of living rooms contain at least one TV. The living room will be matched while the TV will be detected.
As of now, the Detect and Match and Detect options are available only in the US DC.

Training data

For both kinds of entities(labels and objects), Zia has a built-in Gallery with folders containing sample images. These cover a few standard categories, such as:
  1. Bicycles, Motorbikes, Food, and Cars for match
  2. Laptop, Person, Chair, Sofa, and so on for detect
For most use cases, you will need to train Zia using your own images. These are called Custom Images. You will need at least five images for each entity (label or object) that you wish to detect.

In addition to adding images in this manner, you can also:
  1. Add images that have been suggested for training by reviewers (depending on your rule's configuration).
  2. Remove images from training data.
You can have a maximum of 300 custom images for a validation rule. This includes the uploaded images, suggested images, and approved images.

Once there has been a change of at least 40% in the training data (including both additions and removals), Zia will retrain herself and start validating images based on the updated training data.

Metrics

Success rate

The success rate measures how accurate Zia's image validation has been. It compares the number of accurately validated images to the total number of validated images. It is calculated periodically by Zia and is reset when the model is retrained.

Zia can make two kinds of mistakes:
  1. An undesired image is approved by Zia: In this case, the user can remove the image.
  2. A desired image has been rejected by Zia: In this case, the reviewer can approve the image.
In both these cases, the users will have the option of letting Zia know that a mistake has been made. This will reduce the success rate.

If an image is allowed to stay in the system, Zia understands this to be a successful validation. If an image has been removed or approved but the option to notify Zia is not used, then Zia understands those images to be successful validations as well. The success rate will remain the same.

Training accuracy or Model score

The training accuracy or model score reflects the quality of your training data. The closer the training images are to the specified guidelines, the higher its training accuracy/model score. If you have multiple labels or objects, you'll be able to see the training accuracy score for each of them. Only rules with a training accuracy of over 80% will be used for validation.

Where can I find Zia Vision?

Users with the requisite permissions in their profiles will be able to access Zia Vision at Setup () > Zia > Vision.
The following features are available in your Image Validation tab:

  1. New Rule: Use this button to create a new image validation rule.
  2. Image Validation Rules table: Use this to manage all your image validation rules. You will also be able to see useful details like the following:
    1. The module in which a rule is applied. You can filter the rules by module.
    2. The layout in which a rule is applied.
    3. The image field that is validated by the rule.
    4. The process used for validation (Match only, Detect only, Match and detect).
    5. The Model score which represents the quality of your training data. The rule will only be applied if its model score is above 80%.
    6. The status of the rule. You can filter by the statuses (All status, Active, Inactive).
  3. Activate or deactivate a rule: Use this to toggle the status of each rule. Only active rules will be used to validate images.

To create an image validation rule

  1. Navigate to Setup > Zia > Vision > Get Started.
    If you have created rules before, click New Rule.
  2. In the Create Image Validation Rule page, enter the Rule name.
  3. Under the Where to validate section,
    1. Select the module from the drop-down list. This is the module where the image field is located.
    2. Select the required layout. For example, you could choose the Standard layout.
    3. Select the image field that needs to be validated.
      You can either validate the record image, or a custom image upload field in a module.
    4. Set the criteria. This could be either 'All records' or 'Selected records'. In case of 'Selected records', only images in records that satisfy that criteria will be validated by the rule.
  4. Under the Validation Type section, select the type of validation you need for that image. This could be Match only, Detect only, or Match and detect.
  5. The options available in the Upload Training Data section will depend on the validation type chosen in the previous step. For each label or object, you can:
    1. Provide a name
    2. Choose whether the training images represent desired or undesired images in the case of labels. For each object, you need to choose if the images represent an object that needs to be detected or undetected.
    3. Add the training data from the Gallery or from your local device. If you are uploading from your local device:
      1. Ensure that the images have been moved or copied to a separate folder.
      2. Zip that folder.
      3. Upload the zipped folder.
    4. In case you are uploading images from your local device, you'll be able to:
      1. Add multiple labels or objects.
      2. Upload multiple folders to training images for each label or object.
        To learn more about this, refer to the section on uploading training data in this help document.
  6. Under How would you like to add images to feedback learning?, you can decide what happens when an image rejected by Zia is approved by a reviewer. Select one of the following options:
    1. No, feedback learning is not needed: Choose this when you don't want Zia to be trained on images where it has made a mistake.
    2. Suggestions manually provided by Reviewers to users who manage rules, followed by the users' approval(s) to the suggestions accordingly: Reviewers will manually select the images to be added as suggestions for training. Users who manage rules will then pick from these suggestions and add them to the training set.
    3. Suggestions automatically provided by Reviewers to users who manage rules, followed by the users' approval(s) to the suggestions accordingly: Images that are approved by reviewers will be added as suggestions automatically. The admin will need to pick from these suggestions and add them to the training set.
  7. Click Save.
Note
  1. You can set an image validation rule for both standard and custom modules.
  2. Along with record images, only custom image fields with the Maximum images allowed property set to 1 are available for validation.
  3. Whenever an image fails validation during record creation, the image will be sent for manual approval. The associated record will be created.
    The image will wait under the My Jobs module to be approved or rejected by your reviewers.
  4. Sometimes, you may feel that an image has been incorrectly approved by Zia. In those cases, you can remove the image from the record. When you do so, don't forget to select the checkbox in the popup that appears. This will help in producing an accurate success rate for the model.
  5. For a single image field, only one rule can be configured.

To test your image validation rule

Once the rule has been created, you can test the model that Zia will be using to validate incoming images. Based on this testing, you can tweak the model by modifying your training images. To test a model:
  1. Navigate to Setup > Zia > Vision.
  2. Select the rule for which you want to test the model.
  3. Click Test the model.
  4. In the Test the model popup, click browse or drag a file onto the popup to upload the image that needs to be validated by Zia.
    Please note the supported formats and size limit mentioned in the popup before uploading your image.
  5. Zia will validate the image and deliver the result.
  6. Click Test another image to repeat the process for another image.

To review invalid images

When an image is invalidated by Zia, it is moved to the My Jobs module for manual review. To review these images:
  1. Navigate to My Jobs > Image Validation.
  2. For the record whose images you want to review, click the number of images waiting link.

  3. In the Image Validation Failure popup, click:
    1. Accept if image is valid.
    2. Remove if the image is invalid.
  4. You can help Zia by pointing out if Zia made a mistake and/or by suggesting that image for training. Please note that this suggestion option will only be available if you've chosen to let reviewers manually suggest images for feedback learning.
  5. Click Save & Close.

To update an image validation rule

  1. Navigate to Setup > Zia > Vision.
  2. Hover over the rule that you want to edit.
  3. Hover over the (...) icon and select Edit.
  4. Make the necessary changes.
  5. Click Save.
Note
  1. You will not be able to modify the Module, Layout, Field, and Validation Type associated with that rule.
  2. You will be able to modify the Rule Name, Criteria for records, Training Data, and Feedback type.
  3. You will also be able to edit the rule by selecting a rule and clicking the Edit button.

To delete an image validation rule

  1. Navigate to Setup > Zia > Vision.
  2. Hover over the rule that you want to delete.
  3. Hover over the (...) icon and select Delete.
  4. In the popup that appears, click Yes, Delete.
Note
You will also be able to delete a rule by selecting a rule and clicking the Delete button.

Managing your training data

You'll need to add training data during the creation of an image validation rule, as well as when you feel that the performance of the rule could be improved.

To add your training data

Based on the validation type chosen for an image validation rule, you'll have different options under the Upload Training Data section.

If Match only is selected
  1. In the Upload Training Data section, select
    1. Desired if you want your records' images to match your training images. For example, if you have beautiful photographs of properties and you want your property records' images to match those.
    2. Undesired if you want your records' images to be unlike your training images.
      For example, if you find that users sometimes upload images of the interior of a property as records' images. In this case, you can upload multiple images of the interiors of properties and select Undesired.
  2. Enter the label name for the set of training images that you're going to upload. For example, if you're uploading images of houses, then the label would be 'House'.
  3. Click Upload Image.
  4. In the Upload Training Data popup, you can either:
    1. Select the Gallery tab and select one of the available folders.
    2. Select the Desktop tab and upload one or more zipped folders containing custom images. For example, one zipped folder could contain images of the exterior of houses while the other could contain the same but for duplex units.
  5. Click Attach.
Note (only applicable in cases where you upload custom images)
  1. You can add additional custom images for a label by clicking +Images.
  2. You can add a maximum of three labels. To do this, click Add another label.
  3. Each validation rule will have one desired/undesired option. This will apply to all labels in that rule.
  4. Since labels are for entities that occupy at least 80% of the image, there can only be one label detected in an image. If multiple labels are present in a rule, the criteria pattern for the rule will always be Label 1 OR Label 2 OR Label 3. This cannot be modified.
If Detect only is selected
  1. Under the Upload Training Data section, enter the name of the object (Object name) in the training data.
  2. Click Upload Image.
  3. In the Upload Training Data popup, you can either:
    1. Select the Gallery tab and select one of the available folders.
    2. Select the Desktop tab and upload one or more zipped folders containing custom images of the object.
  4. Click Attach.
  5. Decide if you want the object to be detected or undetected.
  6. Click Add another object if you want to validate the presence or absence of other objects in an image. This option is only available when you upload custom images. You can add a maximum of three objects.
  7. If you have multiple objects, set the criteria for validation as well. For example, you may want the image to contain either a television or a combination of a table and a vase. In that case, click Edit Pattern, enter the criteria pattern, and click the tick icon.

Note
You need the specific object to be crystal clear in every image you upload as part of the training data. For example, if you want to upload images of sofas, ensure that each training image only contains a sofa and nothing else. If the object will be present in multiple angles in different records' images, provide training images of the object in as many angles as possible.

The following are valid training images for detecting a sofa.
The image shown below is not recommended for detecting a sofa.

If Match and detect is selected
The match and detect option allows you to match an image as well as detect objects within it. For example, you may want to detect to ensure that images of living rooms are:
  1. Matched with high quality images of living rooms
  2. Also validated by detecting the presence of a TV, vase, and table in that image
To do this:
  1. Configure the Match section and add images for that section. Follow the instructions given for the Match only option.
  2. Repeat the same for the Detect section. Follow the instructions given for the Detect only option.

To add or remove individual images from your training data

This can be done only in the case of custom images being used for training.
  1. Navigate to Setup > Zia > Vision.
  2. Select the rule whose training data you wish to view.
  3. Under the Snapshot of Training Images section, you can view the complete breakdown of you training data. Click View Images.
  4. You can see the following sections:
  1. Images added for training: These are the images that you've uploaded, or added via feedback learning. You can filter and delete images in this section.
  2. Suggestions for feedback learning: These images have been suggested for feedback learning. This is available if you've enabled either manual or automatic suggestions in your image validation rule. You can hover over an image, then approve or reject those images.
    Upon approval, you can select the label from the dropdown list.
    You can then click Crop and Assign Objects, draw a box around your object, and select the object name. You can do this for multiple objects. Next, click Add to Learning.
  3. Approved images for learning: This is the section that holds the images that you've approved in the Suggestions for feedback learning section. You can filter and delete images from this section.

Note
If the training images are from the Gallery, you will not be able to suggest and add training images. You will be able to view sample images from the Gallery.

How Zia updates your model when you add and remove training images

Zia will retrain the model in the case of two events:
  1. When uploaded custom image folders are added or removed and the rule is saved
  2. When there has been a change of 40% and above in the training data. Let's say that the initial training set has 100 images. Over time, you've done the following:
    1. Deleted 10 images from the Images added for training section (10 changes)
    2. Approved 10 suggested images for training (20 changes)
    3. Approved another 10 images (10 changes). Since the total number of changes is now 40 (10+20+10), the change in the training set is 40% when compared to the number of images in the initial training set (40 changes in an initial training set of 100 images). Although only 30 additional images have been added, Zia will retrain the model as there have been 40 changes (30 addition and 10 deletion).
Note
The images in the Suggestions for feedback learning are added by your reviewers either manually or automatically. They can:
  1. Approve images that Zia's rejected in the My Jobs module.
  2. Reject images that Zia's approved by removing that image from the record.

Image upload guidelines

If you want to upload images from your desktop, you must follow these guidelines to achieve the best results from Zia:
  1. The images must be in these formats: JPG, JPEG, PNG, GIF, BMP, and TIFF.
  2. The training data should be nearly similar to the data that needs to be validated. That is, images of villas, motorbikes, cars etc. should be clear and easily identifiable for accurate validation.
  3. In general, the training data should have images from multiple angles, resolutions, and backgrounds for variety.
  4. Vision models generally cannot recognize patterns that humans cannot. So if a human cannot recognize a pattern by looking at the image for around half to one second, the model probably cannot be trained to do it either.
  5. You can upload as many images as you want for better accuracy, but a minimum five images of a particular category must be uploaded for Zia to validate images.
  6. Do not combine images from different categories. Use images that best depict your category.
  7. The training accuracy score will be impacted if these guidelines are not followed.
  8. We recommend you to select record or image approval as the action until you are sure of the results shown by Zia.

Common issues and their solutions

Take a look at the cases below that highlight possible problems and the ways to troubleshoot them:  

  1. Case 1. A stationary supplier wants to validate images of pens and pencils received through bulk orders from the vendors. They upload the images of these objects as classifiers. However, when bulk orders are received, most images of ballpoint pens are marked invalid.

    Solution: The training data consists of images of all kinds of pens, such as fountain pens, gel pens, marker pens, and pencils, but does not include ballpoint pens. Therefore, although Zia can identify a variety of pens, it cannot identify a ballpoint pen and shows it as an invalid image. To get an accurate result, it is important to upload all images of all the objects that you want to be classified.

  2. Case 2. A mobile phone dealer wants to validate images of smart phones. They upload pictures of all the brands of smartphones they sell, but still some images of mobile phones are shown as invalid.

    Solution: The dealer uploads front images of all brands of smartphones. However, some images that are taken from other angles are difficult to identify and draw similarity between objects. Therefore, it is essential to upload images of objects from various angles for ease of identification and accuracy of results.
  3. Case 3. A distributor of toys and puzzles is confused because images of geometric or shape puzzles are identified as invalid objects, even though the same type of images have been used as classifiers.

    Solution: A geometric shape appears similar from every angle, so uploading an image of an object from either angle would always give accurate results. However, there must be enough images to draw similarity between objects. While uploading images of objects that may appear similar from all angles it is necessary to have at least five to six images (minimum requirement for Zia's image validation) of the object. We recommend repeating one image five times for the system to draw a correlation.



    Access your files securely from anywhere

      Zoho CRM Training Programs

      Learn how to use the best tools for sales force automation and better customer engagement from Zoho's implementation specialists.

      Zoho CRM Training
        Redefine the way you work
        with Zoho Workplace

          Zoho DataPrep Personalized Demo

          If you'd like a personalized walk-through of our data preparation tool, please request a demo and we'll be happy to show you how to get the best out of Zoho DataPrep.

          Zoho CRM Training

            Create, share, and deliver

            beautiful slides from anywhere.

            Get Started Now


              Zoho Sign now offers specialized one-on-one training for both administrators and developers.

              BOOK A SESSION









                                            You are currently viewing the help pages of Qntrl’s earlier version. Click here to view our latest version—Qntrl 3.0's help articles.




                                                Manage your brands on social media

                                                  Zoho Desk Resources

                                                  • Desk Community Learning Series


                                                  • Digest


                                                  • Functions


                                                  • Meetups


                                                  • Kbase


                                                  • Resources


                                                  • Glossary


                                                  • Desk Marketplace


                                                  • MVP Corner


                                                  • Word of the Day


                                                    Zoho Marketing Automation

                                                      Zoho Sheet Resources

                                                       

                                                          Zoho Forms Resources


                                                            Secure your business
                                                            communication with Zoho Mail


                                                            Mail on the move with
                                                            Zoho Mail mobile application

                                                              Stay on top of your schedule
                                                              at all times


                                                              Carry your calendar with you
                                                              Anytime, anywhere




                                                                    Zoho Sign Resources

                                                                      Sign, Paperless!

                                                                      Sign and send business documents on the go!

                                                                      Get Started Now




                                                                              Zoho TeamInbox Resources



                                                                                      Zoho DataPrep Resources



                                                                                        Zoho DataPrep Demo

                                                                                        Get a personalized demo or POC

                                                                                        REGISTER NOW


                                                                                          Design. Discuss. Deliver.

                                                                                          Create visually engaging stories with Zoho Show.

                                                                                          Get Started Now









                                                                                                              • Related Articles

                                                                                                              • Capabilities of Zia in Zoho CRM— A perspective

                                                                                                                Zoho CRM harnesses the transformative power of Artificial Intelligence (AI) to revolutionise decision-making across all critical areas of your business, from initial customer interactions to their entire journey. With Zia, AI-powered assistant, Zoho ...
                                                                                                              • Duplicate Image Detection

                                                                                                                Availability Enterprise and Ultimate This feature is currently available in US DC for organizations with more than 20 users. The Duplicate image detection feature aims to assist users in identifying duplicate facial images within a module. This ...
                                                                                                              • Zia Recommendation

                                                                                                                Recommendation tool and its benefits A recommendation tool uses artificial intelligence to identify and analyze customer data such as their purchase details, interests, requirements, and behavioral patterns to suggest the most relevant product. It ...
                                                                                                              • Working with Validation Rules

                                                                                                                Creating and keeping high-quality data is difficult, but it's even harder to identify and fix a dataset full of mistakes. Faulty CRM data can lead to negative consequences for a company, including impeding operations, generating unreliable reports, ...
                                                                                                              • Zia Field Prediction

                                                                                                                Zia's field prediction builder is a toolkit for CRM administrators to build custom predictions in their business. This simple and intuitive builder can quickly predict probable outcomes for both standard and custom modules. What can you predict? You ...
                                                                                                                Wherever you are is as good as
                                                                                                                your workplace

                                                                                                                  Resources

                                                                                                                  Videos

                                                                                                                  Watch comprehensive videos on features and other important topics that will help you master Zoho CRM.



                                                                                                                  eBooks

                                                                                                                  Download free eBooks and access a range of topics to get deeper insight on successfully using Zoho CRM.



                                                                                                                  Webinars

                                                                                                                  Sign up for our webinars and learn the Zoho CRM basics, from customization to sales force automation and more.



                                                                                                                  CRM Tips

                                                                                                                  Make the most of Zoho CRM with these useful tips.



                                                                                                                    Zoho Show Resources