Getting started with IRIS+ Insights

17min

IRIS+ Insights is a brand new module that Irisity offers for customers who want more curated insight into the trends and statistics from their monitoring sites.

Introduction to Insights

The Insights module is a new one that offers customers a perspective across the detectable activities at their sites. These data sets are taken from occurrences triggered in the past, and collected to allow customers to splice the data to their desired visualization preferences.

The Insights module contains these primary functions

  • Create and personalize your dashboards
  • View and make sense of a dashboard

Before you begin!

To ensure that your dashboards generate properly, please make sure you meet the minimum requirements:

  • Have an active camera on an edge device with one of the following rules or statistics active:
    • Statistic rules
      • Crowd statistics
      • People counting
      • Traffic statistics
    • Any real-time event rules
  • At least one event or one statistic reported

The Insights module relies on many functions, each one renders a specific part of the dashboard in a particular way. Therefore, it is crucial to know the general fields that will most likely be used regularly when creating a dashboard.

Data sources

A data source is essentially the back-end table that holds all instances of real-time events or statistics. Each data source provides a list of data that describes events, which typically includes:

  • Event reported start & end time
  • Rule type
  • Object detected
  • Time of event or statistic reported
  • Camera name
  • Camera ID
  • Folder information

You must select the right data source that includes the actual data you need. To do so, be sure to remember which cameras have the rules or statistics you are interested in. Selecting the incorrect data sources of the wrong cameras will make your panels blank!

Graph Types

There are 11 types of graphs available in Insights. This includes bar charts, pie charts, pivot tables, grid tables, etc...the most popular graphs to use are

  • Pivot / Grid table
  • Graph chart
  • Pie chart
  • Cards
Document image


A grid table will typically look like an Excel sheet, where the table consists of columns and rows of information. Each column has a header, each row has an interval of data.

Document image


A graph chart represents bars in a graph, which contains an x-axis and a y-axis.

Document image


A pie chart represents a circle divided into pieces, hence the name "pie". Each piece represents a portion of the entirety of a source.

Document image


A card represents one value taken from a source. This could be taken from any source specified from your data source.



Document image


Bindings

To render a visual in a panel of a chart, you must bind fields from a database to show. There are a few places you must have bindings for in each panel:

  • Values
  • Arguments
  • Series

For particular types of panels there may be additional points of information you can also configure. To learn more, check out the advanced user guide

The Insights module can offer quick and easy visualization functions to help you get a dashboard up and running without any hassle. The most basic visualization options the module offers are:

  • Creating an aggregation of data over time in a graph
  • Creating a count in a panel that updates every couple of minutes
  • Creating columns of fields of interest

To create a new dashboard, click on the button: +Create a new dashboard.



Document image


This will prompt you to create a new dashboard by giving it a name followed by an initial data source to which to connect the dashboard.

Document image


Today, we offer the following types of data sources to connect with your dashboards:

  • IRIS+ Crowd statistics rules: This selects data from cameras that collect crowd statistics
  • IRIS+ People counting rules: This selects data from cameras that collect people counting statistics
  • IRIS+ Traffic statistics rules: This selects data from cameras that collect traffic statistics
  • IRIS+ Health events: This selects data from cameras that have generated health events or alerts
  • IRIS+ Events: This selects data from cameras that have generated real-time events

These data sources are derived from all histories of events and statistics that have occurred from your cameras and devices within the past 30 days.

To properly generate data in the dashboards, it is recommended that you select a data source with cameras configured with the same statistics or events for at least a minute.

You can connect your dashboard to more data sources from this list after its creation.

Examples of Insight dashboards

Below you can find a list of example dashboards that can help you get started:

To create a basic crowd statistics dashboard, create a graph by selecting the following functions in order:

  1. Click on the graph icon to create a graph panel

    Document image
    
  2. Click on the gear icon to open the binding options. To bind, set up the following fields
    1. Add values - Average crowd count
    2. Arguments - End time
    3. Add series - Camera name
Document image


Be sure to click Save at this point to lock in the fields you have just set up.

Lastly, go to the time range selector and select a time range that best fits the duration of time you need. In this example, we will use the range Last 7 days to show the average crowd count across each camera for the last 7 days.

To create a basic object-counting dashboard, create a graph by Clicking on the same graph chart option. Then, select the gear icon and configure the following fields

Document image


Once completed, your graph should look similar to this over the course of a 30 day range

Document image


Example: Basic vehicle statistics dashboard in a graph chart

To create a basic vehicle statistics dashboard, select IRIS+ Traffic Statistics as your data source first. Then, click on the same graph panel option, followed by the gear icon when hovering on the blank panel.

Then, set the binding fields as follows:

  • Add Values. Traffic type (Count)
  • Add argument: Stop time, with group interval: Week-Year
  • Series: Traffic type
Document image


This creates a view of a graph that showcases the types of vehicles detected across cameras by the week. The date displayed on the x-axis shows the date that starts at the beginning of the week.

Document image




Example: Show the latest value of events created in card chart

Start by creating a new card panel with the data source IRIS+ events selected

Document image


Then set the fields in the following manner:

Document image


This will then create a view of cards based on the rules name set, showing both the latest count reported versus a show of count over time.

Document image


Top left number shows the current count reported, middle label shows the rule name, and lastly the sparkline is showing the counts reported overtime.

Example: Basic pivot table of crowd statistics

To create a basic pivot table with the crowd statistics data source, select the pivot table option

Document image


Then, set the fields in the same manner shown here

Document image


Once you have set this up, the pivot table should start populating information if there are any available

Document image


Each account will have a default Insights dashboard, created by us to allow you to play and interact with a dashboard for the first time. This is great for customers who are new to BI tools and perfect for those who wish to learn how they can take advantage of data visualization tools.

To start, locate the Insights tab in your account. If you do not have access to one, contact your sysadmin or Irisity customer support to learn more.

Document image


In this Insights tab, you will see a default Insights dashboard that has already been created:

Document image


This dashboard can be customized and manipulated to any degree, providing flexibility for those learning the functions for the first time.

For those that have created a new dashboard, you can access a saved dashboard by clicking on the Open dashboard button at the top of the dashboard options. This opens a menu list of existing dashboards you can load.

Document image


Start by clicking on Edit dashboard after creating a new dashboard.

Document image


Be sure to click Save after you have selected 10 or less cameras in the camera tree. Once this is done, click on Edit dashboard to open up the edit mode.



Document image


In the example above, you can see that we have selected all the dashboards under the folder "Statistics".

You should also now see options on the left filled with different visualization options. The most common options we offer are

  • Graph
  • Grid
  • Sum
  • Count panel
  • Pie chart
  • Scatter chart
Document image


Other miscellaneous options include

  • Uploading an image
  • Adding text boxes

Each panel will require you to configure at least two fields

  • Values
  • Arguments

These two fields will set one set of data against another. If you've ever created a graph in excel, this is a familiar function.

Document image


In the values section: Click Add Values

  • This will open up dialogue to select and available value type. This value type is dependent to the data source(s) you have selected or added to your dashboard.
  • In this example, the data source containing event information has been selected, because in my case I want to see the count of all events that have occurred across time.

    Document image
    
  • As an argument field, we will now select the argument as camera, because we would like to see the different numbers of events that occurred between different cameras overtime.
  • To visualize metric over time, you will need to set at least the argument field to start time, and specify showing the trend per specific time interval



Document image


Once you have applied the argument field, you will now see your first panel populated with the value and fields configured across a duration of time:

Document image


To add a new panel in the same dashboard, simply click on another visualization option and the dashboard will automatically place and size the blank panel in the same view, running through the same exercise we just did.

Document image


But what if you want to add another data source in the same view?



Document image


Managing a dashboard's layout

Managing a dashboard's layout is very simple. All you will need to do in the edit mode is to drag and drop each panel by selecting each panel at a time and moving them around the dasbboard:

Document image


Dropping them in the screen will automatically scale the panel to fill the page

Document image


What if I want to see trends and statistics across all of my cameras instead of just 10 at a time?

We are currently working on supporting an all-camera single dashboard view, for now please create separate dashboards for each 10 cameras selected.

Do you currently support dashboards for real time data?

For now only events as a count is supported in the Insights module.

Can I have multiple dashboards brought into the same screen at the same time?

We currently do not support multi-dashboard views, for now a dashboard can be shown one at a time.

I see that other options and modes are available in the dashboard, do you have additional guides for those functions?

We are currently working on rolling out more guides and educational content around these functions.