This is the official manual for the latest Org-mode release.
You can label and caption various category of objects—an inline image, a
table, a LaTeX fragment or a Math formula—using
#+CAPTION lines. See Images and tables. ODT exporter enumerates
each labeled or captioned object of a given category separately. As a
result, each such object is assigned a sequence number based on order of its
appearance in the Org file.
In the exported document, a user-provided caption is augmented with the category and sequence number. Consider the following inline image in an Org file.
#+CAPTION: Bell curve #+LABEL: fig:SED-HR4049 [[./img/a.png]]
It could be rendered as shown below in the exported document.
Figure 2: Bell curve
You can modify the category component of the caption by customizing the
org-odt-category-map-alist. For example, to tag all embedded
images with the string ‘Illustration’ (instead of the default
‘Figure’) use the following setting:
(setq org-odt-category-map-alist (("__Figure__" "Illustration" "value" "Figure" org-odt--enumerable-image-p)))
With this, previous image will be captioned as below in the exported document.
Illustration 2: Bell curve