meta elements aren't special, you can query for them and get their attributes in the normal way.
In this case, here's how you'd get the content attribute value from the first meta[property="og:image"] element:
var element = document.querySelector('meta[property~="og:image"]');
var content = element && element.getAttribute("content");
querySelector is supported by all modern browsers, and also IE8.
Note that the content property is also available as a reflected property, so you can just use .content rather than .getAttribute("content"):
var element = document.querySelector('meta[property~="og:image"]');
var content = element && element.content;
In modern JavaScript you can use the optional chaining operator (?.) to combine those two statements:
const content = document.querySelector('meta[property~="og:image"]')?.content;
// −−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−^
If the element isn't found, content will get the value undefined; otherwise, it'll get the value of the reflected property (which is the attribute value).
<meta>is an element,propertyis an attribute. you'd want something more likemeta = document.getElementsByTagName('meta'), scan through that for your og:image attribute, then get the associated content.