C/C++ - Thermal Imaging Bricklet

This is the description of the C/C++ API bindings for the Thermal Imaging Bricklet. General information and technical specifications for the Thermal Imaging Bricklet are summarized in its hardware description.

An installation guide for the C/C++ API bindings is part of their general description.

Examples

The example code below is Public Domain (CC0 1.0).

Callback

Download (example_callback.c)

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
#include <stdio.h>

#include "ip_connection.h"
#include "bricklet_thermal_imaging.h"

#define HOST "localhost"
#define PORT 4223
#define UID "XYZ" // Change XYZ to the UID of your Thermal Imaging Bricklet

// Callback function for high contrast image callback
void cb_high_contrast_image(uint8_t *image, uint16_t image_length, void *user_data) {
    (void)image; (void)image_length; (void)user_data; // avoid unused parameter warning

    // image is a array of size 80*60 with 8 bit grey value for each element
}

int main(void) {
    // Create IP connection
    IPConnection ipcon;
    ipcon_create(&ipcon);

    // Create device object
    ThermalImaging ti;
    thermal_imaging_create(&ti, UID, &ipcon);

    // Connect to brickd
    if(ipcon_connect(&ipcon, HOST, PORT) < 0) {
        fprintf(stderr, "Could not connect\n");
        return 1;
    }
    // Don't use device before ipcon is connected

    // Register high contrast image callback to function cb_high_contrast_image
    thermal_imaging_register_callback(&ti,
                                      THERMAL_IMAGING_CALLBACK_HIGH_CONTRAST_IMAGE,
                                      (void *)cb_high_contrast_image,
                                      NULL);

    // Enable high contrast image transfer for callback
    thermal_imaging_set_image_transfer_config(&ti,
                                              THERMAL_IMAGING_IMAGE_TRANSFER_CALLBACK_HIGH_CONTRAST_IMAGE);

    printf("Press key to exit\n");
    getchar();
    thermal_imaging_destroy(&ti);
    ipcon_destroy(&ipcon); // Calls ipcon_disconnect internally
    return 0;
}

API

Every function of the C/C++ bindings returns an integer which describes an error code. Data returned from the device, when a getter is called, is handled via call by reference. These parameters are labeled with the ret_ prefix.

Possible error codes are:

  • E_OK = 0
  • E_TIMEOUT = -1
  • E_NO_STREAM_SOCKET = -2
  • E_HOSTNAME_INVALID = -3
  • E_NO_CONNECT = -4
  • E_NO_THREAD = -5
  • E_NOT_ADDED = -6 (unused since bindings version 2.0.0)
  • E_ALREADY_CONNECTED = -7
  • E_NOT_CONNECTED = -8
  • E_INVALID_PARAMETER = -9
  • E_NOT_SUPPORTED = -10
  • E_UNKNOWN_ERROR_CODE = -11
  • E_STREAM_OUT_OF_SYNC = -12

as defined in ip_connection.h.

All functions listed below are thread-safe.

Basic Functions

void thermal_imaging_create(ThermalImaging *thermal_imaging, const char *uid, IPConnection *ipcon)

Creates the device object thermal_imaging with the unique device ID uid and adds it to the IPConnection ipcon:

ThermalImaging thermal_imaging;
thermal_imaging_create(&thermal_imaging, "YOUR_DEVICE_UID", &ipcon);

This device object can be used after the IP connection has been connected (see examples above).

void thermal_imaging_destroy(ThermalImaging *thermal_imaging)

Removes the device object thermal_imaging from its IPConnection and destroys it. The device object cannot be used anymore afterwards.

int thermal_imaging_get_high_contrast_image(ThermalImaging *thermal_imaging, uint8_t *ret_image, uint16_t *ret_image_length)

Returns the current high contrast image. See here for the difference between High Contrast and Temperature Image. If you don't know what to use the High Contrast Image is probably right for you.

The data is organized as a 8-bit value 80x60 pixel matrix linearized in a one-dimensional array. The data is arranged line by line from top left to bottom right.

Each 8-bit value represents one gray-scale image pixel that can directly be shown to a user on a display.

Before you can use this function you have to enable it with thermal_imaging_set_image_transfer_config().

int thermal_imaging_get_temperature_image(ThermalImaging *thermal_imaging, uint16_t *ret_image, uint16_t *ret_image_length)

Returns the current temperature image. See here for the difference between High Contrast and Temperature Image. If you don't know what to use the High Contrast Image is probably right for you.

The data is organized as a 16-bit value 80x60 pixel matrix linearized in a one-dimensional array. The data is arranged line by line from top left to bottom right.

Each 16-bit value represents one temperature measurement in either Kelvin/10 or Kelvin/100 (depending on the resolution set with:c:func:thermal_imaging_set_resolution).

Before you can use this function you have to enable it with thermal_imaging_set_image_transfer_config().

int thermal_imaging_get_statistics(ThermalImaging *thermal_imaging, uint16_t ret_spotmeter_statistics[4], uint16_t ret_temperatures[4], uint8_t *ret_resolution, uint8_t *ret_ffc_status, bool ret_temperature_warning[2])

Returns the spotmeter statistics, various temperatures, current resolution and status bits.

The spotmeter statistics are:

  • Index 0: Mean Temperature.
  • Index 1: Maximum Temperature.
  • Index 2: Minimum Temperature.
  • Index 3: Pixel Count of spotmeter region of interest.

The temperatures are:

  • Index 0: Focal Plain Array temperature.
  • Index 1: Focal Plain Array temperature at last FFC (Flat Field Correction).
  • Index 2: Housing temperature.
  • Index 3: Housing temperature at last FFC.

The resolution is either 0 to 6553 Kelvin or 0 to 655 Kelvin. If the resolution is the former, the temperatures are in Kelvin/10, if it is the latter the temperatures are in Kelvin/100.

FFC (Flat Field Correction) Status:

  • FFC Never Commanded: Only seen on startup before first FFC.
  • FFC Imminent: This state is entered 2 seconds prior to initiating FFC.
  • FFC In Progress: Flat field correction is started (shutter moves in front of lens and back). Takes about 1 second.
  • FFC Complete: Shutter is in waiting position again, FFC done.

Temperature warning bits:

  • Index 0: Shutter lockout (if true shutter is locked out because temperature is outside -10°C to +65°C)
  • Index 1: Overtemperature shut down imminent (goes true 10 seconds before shutdown)

The following defines are available for this function:

  • THERMAL_IMAGING_RESOLUTION_0_TO_6553_KELVIN = 0
  • THERMAL_IMAGING_RESOLUTION_0_TO_655_KELVIN = 1
  • THERMAL_IMAGING_FFC_STATUS_NEVER_COMMANDED = 0
  • THERMAL_IMAGING_FFC_STATUS_IMMINENT = 1
  • THERMAL_IMAGING_FFC_STATUS_IN_PROGRESS = 2
  • THERMAL_IMAGING_FFC_STATUS_COMPLETE = 3
int thermal_imaging_set_resolution(ThermalImaging *thermal_imaging, uint8_t resolution)

Sets the resolution. The Thermal Imaging Bricklet can either measure

  • from 0 to 6553 Kelvin (-273.15°C to +6279.85°C) with 0.1°C resolution or
  • from 0 to 655 Kelvin (-273.15°C to +381.85°C) with 0.01°C resolution.

The accuracy is specified for -10°C to 450°C in the first range and -10°C and 140°C in the second range.

The default value is 0 to 655 Kelvin.

The following defines are available for this function:

  • THERMAL_IMAGING_RESOLUTION_0_TO_6553_KELVIN = 0
  • THERMAL_IMAGING_RESOLUTION_0_TO_655_KELVIN = 1
int thermal_imaging_get_resolution(ThermalImaging *thermal_imaging, uint8_t *ret_resolution)

Returns the resolution as set by thermal_imaging_set_resolution().

The following defines are available for this function:

  • THERMAL_IMAGING_RESOLUTION_0_TO_6553_KELVIN = 0
  • THERMAL_IMAGING_RESOLUTION_0_TO_655_KELVIN = 1
int thermal_imaging_set_spotmeter_config(ThermalImaging *thermal_imaging, uint8_t region_of_interest[4])

Sets the spotmeter region of interest. The 4 values are

  • Index 0: Column start (has to be smaller then Column end).
  • Index 1: Row start (has to be smaller then Row end).
  • Index 2: Column end (has to be smaller then 80).
  • Index 3: Row end (has to be smaller then 60).

The spotmeter statistics can be read out with thermal_imaging_get_statistics().

The default region of interest is (39, 29, 40, 30).

int thermal_imaging_get_spotmeter_config(ThermalImaging *thermal_imaging, uint8_t ret_region_of_interest[4])

Returns the spotmeter config as set by thermal_imaging_set_spotmeter_config().

int thermal_imaging_set_high_contrast_config(ThermalImaging *thermal_imaging, uint8_t region_of_interest[4], uint16_t dampening_factor, uint16_t clip_limit[2], uint16_t empty_counts)

Sets the high contrast region of interest, dampening factor, clip limit and empty counts. This config is only used in high contrast mode (see thermal_imaging_set_image_transfer_config()).

The high contrast region of interest consists of four values:

  • Index 0: Column start (has to be smaller or equal then Column end).
  • Index 1: Row start (has to be smaller then Row end).
  • Index 2: Column end (has to be smaller then 80).
  • Index 3: Row end (has to be smaller then 60).

The algorithm to generate the high contrast image is applied to this region.

Dampening Factor: This parameter is the amount of temporal dampening applied to the HEQ (history equalization) transformation function. An IIR filter of the form:

(N / 256) * previous + ((256 - N) / 256) * current

is applied, and the HEQ dampening factor represents the value N in the equation, i.e., a value that applies to the amount of influence the previous HEQ transformation function has on the current function. The lower the value of N the higher the influence of the current video frame whereas the higher the value of N the more influence the previous damped transfer function has.

Clip Limit Index 0 (AGC HEQ Clip Limit Low): This parameter defines an artificial population that is added to every non-empty histogram bin. In other words, if the Clip Limit Low is set to L, a bin with an actual population of X will have an effective population of L + X. Any empty bin that is nearby a populated bin will be given an artificial population of L. The effect of higher values is to provide a more linear transfer function; lower values provide a more non-linear (equalized) transfer function.

Clip Limit Index 1 (AGC HEQ Clip Limit High): This parameter defines the maximum number of pixels allowed to accumulate in any given histogram bin. Any additional pixels in a given bin are clipped. The effect of this parameter is to limit the influence of highly-populated bins on the resulting HEQ transformation function.

Empty Counts: This parameter specifies the maximum number of pixels in a bin that will be interpreted as an empty bin. Histogram bins with this number of pixels or less will be processed as an empty bin.

The default values are

  • Region Of Interest = (0, 0, 79, 59),
  • Dampening Factor = 64,
  • Clip Limit = (4800, 512) and
  • Empty Counts = 2.
int thermal_imaging_get_high_contrast_config(ThermalImaging *thermal_imaging, uint8_t ret_region_of_interest[4], uint16_t *ret_dampening_factor, uint16_t ret_clip_limit[2], uint16_t *ret_empty_counts)

Returns the high contrast config as set by thermal_imaging_set_high_contrast_config().

Advanced Functions

int thermal_imaging_get_api_version(ThermalImaging *thermal_imaging, uint8_t ret_api_version[3])

Returns the version of the API definition (major, minor, revision) implemented by this API bindings. This is neither the release version of this API bindings nor does it tell you anything about the represented Brick or Bricklet.

int thermal_imaging_get_response_expected(ThermalImaging *thermal_imaging, uint8_t function_id, bool *ret_response_expected)

Returns the response expected flag for the function specified by the function ID parameter. It is true if the function is expected to send a response, false otherwise.

For getter functions this is enabled by default and cannot be disabled, because those functions will always send a response. For callback configuration functions it is enabled by default too, but can be disabled by thermal_imaging_set_response_expected(). For setter functions it is disabled by default and can be enabled.

Enabling the response expected flag for a setter function allows to detect timeouts and other error conditions calls of this setter as well. The device will then send a response for this purpose. If this flag is disabled for a setter function then no response is send and errors are silently ignored, because they cannot be detected.

See thermal_imaging_set_response_expected() for the list of function ID defines available for this function.

int thermal_imaging_set_response_expected(ThermalImaging *thermal_imaging, uint8_t function_id, bool response_expected)

Changes the response expected flag of the function specified by the function ID parameter. This flag can only be changed for setter (default value: false) and callback configuration functions (default value: true). For getter functions it is always enabled.

Enabling the response expected flag for a setter function allows to detect timeouts and other error conditions calls of this setter as well. The device will then send a response for this purpose. If this flag is disabled for a setter function then no response is send and errors are silently ignored, because they cannot be detected.

The following function ID defines are available for this function:

  • THERMAL_IMAGING_FUNCTION_SET_RESOLUTION = 4
  • THERMAL_IMAGING_FUNCTION_SET_SPOTMETER_CONFIG = 6
  • THERMAL_IMAGING_FUNCTION_SET_HIGH_CONTRAST_CONFIG = 8
  • THERMAL_IMAGING_FUNCTION_SET_IMAGE_TRANSFER_CONFIG = 10
  • THERMAL_IMAGING_FUNCTION_SET_WRITE_FIRMWARE_POINTER = 237
  • THERMAL_IMAGING_FUNCTION_SET_STATUS_LED_CONFIG = 239
  • THERMAL_IMAGING_FUNCTION_RESET = 243
  • THERMAL_IMAGING_FUNCTION_WRITE_UID = 248
int thermal_imaging_set_response_expected_all(ThermalImaging *thermal_imaging, bool response_expected)

Changes the response expected flag for all setter and callback configuration functions of this device at once.

int thermal_imaging_get_spitfp_error_count(ThermalImaging *thermal_imaging, uint32_t *ret_error_count_ack_checksum, uint32_t *ret_error_count_message_checksum, uint32_t *ret_error_count_frame, uint32_t *ret_error_count_overflow)

Returns the error count for the communication between Brick and Bricklet.

The errors are divided into

  • ACK checksum errors,
  • message checksum errors,
  • framing errors and
  • overflow errors.

The errors counts are for errors that occur on the Bricklet side. All Bricks have a similar function that returns the errors on the Brick side.

int thermal_imaging_set_bootloader_mode(ThermalImaging *thermal_imaging, uint8_t mode, uint8_t *ret_status)

Sets the bootloader mode and returns the status after the requested mode change was instigated.

You can change from bootloader mode to firmware mode and vice versa. A change from bootloader mode to firmware mode will only take place if the entry function, device identifier and CRC are present and correct.

This function is used by Brick Viewer during flashing. It should not be necessary to call it in a normal user program.

The following defines are available for this function:

  • THERMAL_IMAGING_BOOTLOADER_MODE_BOOTLOADER = 0
  • THERMAL_IMAGING_BOOTLOADER_MODE_FIRMWARE = 1
  • THERMAL_IMAGING_BOOTLOADER_MODE_BOOTLOADER_WAIT_FOR_REBOOT = 2
  • THERMAL_IMAGING_BOOTLOADER_MODE_FIRMWARE_WAIT_FOR_REBOOT = 3
  • THERMAL_IMAGING_BOOTLOADER_MODE_FIRMWARE_WAIT_FOR_ERASE_AND_REBOOT = 4
  • THERMAL_IMAGING_BOOTLOADER_STATUS_OK = 0
  • THERMAL_IMAGING_BOOTLOADER_STATUS_INVALID_MODE = 1
  • THERMAL_IMAGING_BOOTLOADER_STATUS_NO_CHANGE = 2
  • THERMAL_IMAGING_BOOTLOADER_STATUS_ENTRY_FUNCTION_NOT_PRESENT = 3
  • THERMAL_IMAGING_BOOTLOADER_STATUS_DEVICE_IDENTIFIER_INCORRECT = 4
  • THERMAL_IMAGING_BOOTLOADER_STATUS_CRC_MISMATCH = 5
int thermal_imaging_get_bootloader_mode(ThermalImaging *thermal_imaging, uint8_t *ret_mode)

Returns the current bootloader mode, see thermal_imaging_set_bootloader_mode().

The following defines are available for this function:

  • THERMAL_IMAGING_BOOTLOADER_MODE_BOOTLOADER = 0
  • THERMAL_IMAGING_BOOTLOADER_MODE_FIRMWARE = 1
  • THERMAL_IMAGING_BOOTLOADER_MODE_BOOTLOADER_WAIT_FOR_REBOOT = 2
  • THERMAL_IMAGING_BOOTLOADER_MODE_FIRMWARE_WAIT_FOR_REBOOT = 3
  • THERMAL_IMAGING_BOOTLOADER_MODE_FIRMWARE_WAIT_FOR_ERASE_AND_REBOOT = 4
int thermal_imaging_set_write_firmware_pointer(ThermalImaging *thermal_imaging, uint32_t pointer)

Sets the firmware pointer for thermal_imaging_write_firmware(). The pointer has to be increased by chunks of size 64. The data is written to flash every 4 chunks (which equals to one page of size 256).

This function is used by Brick Viewer during flashing. It should not be necessary to call it in a normal user program.

int thermal_imaging_write_firmware(ThermalImaging *thermal_imaging, uint8_t data[64], uint8_t *ret_status)

Writes 64 Bytes of firmware at the position as written by thermal_imaging_set_write_firmware_pointer() before. The firmware is written to flash every 4 chunks.

You can only write firmware in bootloader mode.

This function is used by Brick Viewer during flashing. It should not be necessary to call it in a normal user program.

int thermal_imaging_set_status_led_config(ThermalImaging *thermal_imaging, uint8_t config)

Sets the status LED configuration. By default the LED shows communication traffic between Brick and Bricklet, it flickers once for every 10 received data packets.

You can also turn the LED permanently on/off or show a heartbeat.

If the Bricklet is in bootloader mode, the LED is will show heartbeat by default.

The following defines are available for this function:

  • THERMAL_IMAGING_STATUS_LED_CONFIG_OFF = 0
  • THERMAL_IMAGING_STATUS_LED_CONFIG_ON = 1
  • THERMAL_IMAGING_STATUS_LED_CONFIG_SHOW_HEARTBEAT = 2
  • THERMAL_IMAGING_STATUS_LED_CONFIG_SHOW_STATUS = 3
int thermal_imaging_get_status_led_config(ThermalImaging *thermal_imaging, uint8_t *ret_config)

Returns the configuration as set by thermal_imaging_set_status_led_config()

The following defines are available for this function:

  • THERMAL_IMAGING_STATUS_LED_CONFIG_OFF = 0
  • THERMAL_IMAGING_STATUS_LED_CONFIG_ON = 1
  • THERMAL_IMAGING_STATUS_LED_CONFIG_SHOW_HEARTBEAT = 2
  • THERMAL_IMAGING_STATUS_LED_CONFIG_SHOW_STATUS = 3
int thermal_imaging_get_chip_temperature(ThermalImaging *thermal_imaging, int16_t *ret_temperature)

Returns the temperature in °C as measured inside the microcontroller. The value returned is not the ambient temperature!

The temperature is only proportional to the real temperature and it has bad accuracy. Practically it is only useful as an indicator for temperature changes.

int thermal_imaging_reset(ThermalImaging *thermal_imaging)

Calling this function will reset the Bricklet. All configurations will be lost.

After a reset you have to create new device objects, calling functions on the existing ones will result in undefined behavior!

int thermal_imaging_write_uid(ThermalImaging *thermal_imaging, uint32_t uid)

Writes a new UID into flash. If you want to set a new UID you have to decode the Base58 encoded UID string into an integer first.

We recommend that you use Brick Viewer to change the UID.

int thermal_imaging_read_uid(ThermalImaging *thermal_imaging, uint32_t *ret_uid)

Returns the current UID as an integer. Encode as Base58 to get the usual string version.

int thermal_imaging_get_identity(ThermalImaging *thermal_imaging, char ret_uid[8], char ret_connected_uid[8], char *ret_position, uint8_t ret_hardware_version[3], uint8_t ret_firmware_version[3], uint16_t *ret_device_identifier)

Returns the UID, the UID where the Bricklet is connected to, the position, the hardware and firmware version as well as the device identifier.

The position can be 'a', 'b', 'c' or 'd'.

The device identifier numbers can be found here. There is also a constant for the device identifier of this Bricklet.

Callback Configuration Functions

void thermal_imaging_register_callback(ThermalImaging *thermal_imaging, int16_t callback_id, void *function, void *user_data)

Registers the given function with the given callback_id. The user_data will be passed as the last parameter to the function.

The available callback IDs with corresponding function signatures are listed below.

int thermal_imaging_set_image_transfer_config(ThermalImaging *thermal_imaging, uint8_t config)

The necessary bandwidth of this Bricklet is too high to use getter/callback or high contrast/temperature image at the same time. You have to configure the one you want to use, the Bricklet will optimize the internal configuration accordingly.

Corresponding functions:

The default is Manual High Contrast Image (0).

The following defines are available for this function:

  • THERMAL_IMAGING_IMAGE_TRANSFER_MANUAL_HIGH_CONTRAST_IMAGE = 0
  • THERMAL_IMAGING_IMAGE_TRANSFER_MANUAL_TEMPERATURE_IMAGE = 1
  • THERMAL_IMAGING_IMAGE_TRANSFER_CALLBACK_HIGH_CONTRAST_IMAGE = 2
  • THERMAL_IMAGING_IMAGE_TRANSFER_CALLBACK_TEMPERATURE_IMAGE = 3
int thermal_imaging_get_image_transfer_config(ThermalImaging *thermal_imaging, uint8_t *ret_config)

Returns the image transfer config, as set by thermal_imaging_set_image_transfer_config().

The following defines are available for this function:

  • THERMAL_IMAGING_IMAGE_TRANSFER_MANUAL_HIGH_CONTRAST_IMAGE = 0
  • THERMAL_IMAGING_IMAGE_TRANSFER_MANUAL_TEMPERATURE_IMAGE = 1
  • THERMAL_IMAGING_IMAGE_TRANSFER_CALLBACK_HIGH_CONTRAST_IMAGE = 2
  • THERMAL_IMAGING_IMAGE_TRANSFER_CALLBACK_TEMPERATURE_IMAGE = 3

Callbacks

Callbacks can be registered to receive time critical or recurring data from the device. The registration is done with the thermal_imaging_register_callback() function. The parameters consist of the device object, the callback ID, the callback function and optional user data:

void my_callback(int p, void *user_data) {
    printf("parameter: %d\n", p);
}

thermal_imaging_register_callback(&thermal_imaging, THERMAL_IMAGING_CALLBACK_EXAMPLE, (void *)my_callback, NULL);

The available constants with corresponding callback function signatures are described below.

Note

Using callbacks for recurring events is always preferred compared to using getters. It will use less USB bandwidth and the latency will be a lot better, since there is no round trip time.

THERMAL_IMAGING_CALLBACK_HIGH_CONTRAST_IMAGE
void callback(uint8_t *image, uint16_t image_length, void *user_data)

This callback is triggered with every new high contrast image if the transfer image config is configured for high contrast callback (see thermal_imaging_set_image_transfer_config()).

The data is organized as a 8-bit value 80x60 pixel matrix linearized in a one-dimensional array. The data is arranged line by line from top left to bottom right.

Each 8-bit value represents one gray-scale image pixel that can directly be shown to a user on a display.

THERMAL_IMAGING_CALLBACK_TEMPERATURE_IMAGE
void callback(uint16_t *image, uint16_t image_length, void *user_data)

This callback is triggered with every new temperature image if the transfer image config is configured for temperature callback (see thermal_imaging_set_image_transfer_config()).

The data is organized as a 16-bit value 80x60 pixel matrix linearized in a one-dimensional array. The data is arranged line by line from top left to bottom right.

Each 16-bit value represents one temperature measurement in either Kelvin/10 or Kelvin/100 (depending on the resolution set with thermal_imaging_set_resolution()).

Constants

THERMAL_IMAGING_DEVICE_IDENTIFIER

This constant is used to identify a Thermal Imaging Bricklet.

The thermal_imaging_get_identity() function and the IPCON_CALLBACK_ENUMERATE callback of the IP Connection have a device_identifier parameter to specify the Brick's or Bricklet's type.

THERMAL_IMAGING_DEVICE_DISPLAY_NAME

This constant represents the human readable name of a Thermal Imaging Bricklet.