# Image-Based

# ImageToTextTask

Solves image-based captchas by recognizing text from a base64-encoded image.

API type: ImageToTextTask

Parameter	Type	Default	Description
`body`	`str`	required	Base64-encoded image body. Must be without line breaks.
`capMonsterModule`	`str \\| None`	`None`	Recognition module name (e.g. `"yandex"`).
`recognizingThreshold`	`int \\| None`	`None`	Confidence threshold (0–100). You won't be charged if confidence is below this value.
`case`	`bool \\| None`	`None`	Set to `True` if the captcha is case-sensitive.
`numeric`	`Literal[0, 1] \\| None`	`None`	Set to `1` if the captcha contains only numbers.
`math`	`bool \\| None`	`None`	Set to `True` if the captcha requires a math operation.

Solves complex image-based reCAPTCHA challenges (grid selection tasks).

API type: ComplexImageTask (with class: "recaptcha")

Parameter	Type	Default	Description
`imageUrls`	`list[str] \\| None`	`None`	List of image URLs for the captcha parts.
`imagesBase64`	`list[str] \\| None`	`None`	List of base64-encoded images for the captcha parts.
`metadata`	`ComplexImageRecaptchaMetadata`	required	Metadata describing the task.

Exactly one of imageUrls or imagesBase64 must be provided (not both, not neither).

Parameter	Type	Default	Description
`Task`	`str \\| None`	`None`	Textual task description (e.g. `"Click on traffic lights"`).
`TaskDefinition`	`str \\| None`	`None`	Technical task type identifier.
`Grid`	`Literal["4x4", "3x3", "1x1"]`	required	Grid size of the captcha.

Exactly one of Task or TaskDefinition must be set.

Solves complex image recognition tasks (rotate, match, etc.).

API type: ComplexImageTask (with class: "recognition")

Parameter	Type	Default	Description
`imagesBase64`	`list[str]`	required	List of base64-encoded images.
`task`	`Literal["oocl_rotate_new", "oocl_rotate_double_new", "betpunch_3x3_rotate", "bls", "shein"]`	required	Type of recognition task.