Illustrated C/C++ source sheets, symbol index, API relationship graph, compiler console and debug trace

LIVE AI / FLEXIBLE SENSING / FLIGHT CONTEXT

Sensoraia API

Live AI and sensing, connected to the flight.
See, listen, inspect and return supported results while airborne.

Loading...
Searching...
No Matches
lwir_inference.c File Reference

Classifier evidence with explicit display-versus-radiometric model inputs. More...

#include "lwir_inference.h"
#include "lwir_container.h"
#include "lwir_frame.h"
#include "inference_jpeg.h"
#include <math.h>
#include <stdio.h>
#include <string.h>
#include <unistd.h>
Include dependency graph for lwir_inference.c:

Go to the source code of this file.

Functions

bool lwir_inference_open (const struct inference_settings *settings, double threshold, unsigned category, char *error, size_t capacity)
 Open and qualify the LWIR classifier (see lwir_inference.h).
bool lwir_inference_image (const char *filename, unsigned long long sequence, char *error, size_t capacity)
 Classify one saved photo and publish its evidence sidecar.
void lwir_inference_close (void)
 Release the classifier; harmless when not open. Threshold/category are reset by the next open.

Variables

static struct inference_context network
 The single open classifier; storage != NULL means open.
static unsigned char thermal_plane [LWIR_MAX_WIDTH *LWIR_MAX_HEIGHT *2]
 Kelvin*64 samples for radiometric models.
static double score_threshold
 Inclusive acceptance threshold set at open.
static unsigned class_index
 Reported class within the probability vector.

Detailed Description

Classifier evidence with explicit display-versus-radiometric model inputs.

A class probability is neither a temperature correction nor a geographic hotspot result. One model is held in module-static storage (no heap here; the inference context owns its own storage). Pixel-domain models read the display JPEG; other models require the embedded temperature layer. Results go to an exclusively created "<photo>.lwir.ini" sidecar; the photograph itself is never modified.

Definition in file lwir_inference.c.

Function Documentation

◆ lwir_inference_close()

void lwir_inference_close ( void )

Release the classifier; harmless when not open. Threshold/category are reset by the next open.

Release the serialized classifier; repeated closes are harmless.

Definition at line 139 of file lwir_inference.c.

◆ lwir_inference_image()

bool lwir_inference_image ( const char * filename,
unsigned long long sequence,
char * error,
size_t capacity )

Classify one saved photo and publish its evidence sidecar.

Classify one immutable JPEG/temperature plane and exclusively publish its model-identified sidecar.

Returns
True once the sidecar is fully written and fsynced. False if the model is not open, input is missing/invalid, any probability is outside [0,1], hashing fails, or the sidecar exists or cannot be written (a partial sidecar is removed). The photograph is untouched in every case.

Records image and model hashes, input units, synthetic-temperature flag, source size, class, probability and accepted (probability >= threshold).

Note
synthetic_temperature is only meaningful for temperature-input models; pixel-input models always record 0.
The image hash is computed after inference by a separate file read.

Definition at line 83 of file lwir_inference.c.

◆ lwir_inference_open()

bool lwir_inference_open ( const struct inference_settings * settings,
double threshold,
unsigned category,
char * error,
size_t capacity )

Open and qualify the LWIR classifier (see lwir_inference.h).

Open one explicit LWIR classifier with a declared pixel or radiometric input domain.

Returns
True if open (or already open); false on bad arguments, open failure or an unsupported model contract, leaving nothing open.

Requires exactly one input and one float32 output of at most 4096 classes containing category; non-pixel (temperature) inputs must be single channel.

Note
Already open: returns true immediately and keeps the earlier threshold/category.

Definition at line 49 of file lwir_inference.c.

Variable Documentation

◆ class_index

unsigned class_index
static

Reported class within the probability vector.

Definition at line 38 of file lwir_inference.c.

◆ network

struct inference_context network
static

The single open classifier; storage != NULL means open.

Definition at line 35 of file lwir_inference.c.

◆ score_threshold

double score_threshold
static

Inclusive acceptance threshold set at open.

Definition at line 37 of file lwir_inference.c.

◆ thermal_plane

unsigned char thermal_plane[LWIR_MAX_WIDTH *LWIR_MAX_HEIGHT *2]
static

Kelvin*64 samples for radiometric models.

Definition at line 36 of file lwir_inference.c.